Relationships among selected teacher behaviors and characteristics and student perceptions of teacher
warmth, prestige, and effectiveness
by Douglas Nathanial Smith
A thesis submitted in partial fulfillment of the requirements for the degree of DOCTOR OF
EDUCATION
Montana State University
© Copyright by Douglas Nathanial Smith (1977)
Abstract:
The intent of this study was to find out if there was any relationship among selected classroom
behaviors and characteristics of a sample of teachers at Montana State University and student
perceptions of teacher warmth, prestige, and effectiveness. In addition, this study sought to examine the
relationships among these teacher behaviors and characteristics and teacher self-ratings and the grades
the students expected to receive for the course.
Student volunteers were trained in the use of the Teacher Behaviors and Characteristics Checklist, an
instrument developed by the investigator, and then attended three classes of each of the teachers
involved in the study during Spring Quarter 1975 in order to count and categorize teacher behaviors
and characteristics. Three instruments were administered to the students of these classes—the
Authoritativeness Scale, the Scale for Measurement of Counselor Traits, and the Revised Faculty
Rating Form. Data relating to the grades the students expected to receive for the course and teacher
self-ratings as to the effectiveness of their instruction was also collected.
The hypotheses tested in this study were concerned with the relationships among teacher behaviors and
characteristics and five independent measures of teacher performance. These five measures were
teacher warmth, teacher prestige, teacher effectiveness, course value—item 33, and course value—item
85. All hypotheses were tested at either the .05 or the .01 level of significance using either the analysis
of variance or the Pearson correlation.
Results of this study relevant to the measure of warmth indicated that teachers who were perceived by
students as being warmer persons were of higher academic rank, held doctor's degrees, rated
themselves between "average" and "above average" in effectiveness, related more of their positive
experiences to their class, and tended to dress more in ties and dress pants, sports coats and dress pants,
and suits' but did not wear sports coats and suits all of the time. They also positively evaluated student
responses less frequently, did not acknowledge student feelings as often, smiled less, and their students
spoke less in class and expected to receive lower grades for the course.
Teachers who were viewed with more prestige rated themselves between "average" and "above
average" in effectiveness, held doctor's degrees, were expected to give lower grades, and were of
higher academic rank—with one notable exception. Instructors received the highest ratings. Z
RELATIONSHIPS AMONG SELECTED TEACHER BEHAVIORS AND
CHARACTERISTICS AND STUDENT PERCEPTIONS OF TEACHER
WARMTH, PRESTIGE, AND EFFECTIVENESS
by
DOUGLAS NATHANIAL SMITH
A thesis submitted in partial fulfillment
of the requirements for the degree
of
DOCTOR OF EDUCATION
Approved:
Graduate tDean
MONTANA STATE UNIVERSITY
Bozeman, Montana
January, 1977
iii
ACKNOWLEDGMENT
Special acknowledgment is made to my advisor. Dr. Earl Ringo,
for the guidance, encouragement, and support he gave to my study.
Acknowledgment is made to my reading committee, Dr. Albert
Suvak and Dr. Richard Horswill, and to the other members of my
committee. Professor Burl Winchester, Dr. George Rice, and Dr. Ralph
Olsen.
Acknowledgment is also made to Dr. Douglas Herbster, Dr. Joan
Miles, and Dr. Carol Parker for their cooperation in providing me with
the observers needed for the development of my research instrument and ■
the counting and categorizing of teacher behaviors and characteristics.
Special thanks is given Dr. Albert Suvak for assisting me with
the organization and analysis of my data.
'I
TABLE OF CONTENTS
Page
LIST OF TABLES.................................................. vii
ABSTRACT........................................................ ix
Chapter
1.
INTRODUCTION ......................
I
STATEMENT OF THE PROB L E M ..............................
4
PURPOSE OF THE S T U D Y ..................................
4
NEED FOR THE S T U D Y ....................................
5
GENERAL QUESTIONS TO BE INVESTIGATED ................
7
.
GENERAL PROCEDURES ....................................
2.
8
LIMITATIONS OF THE S T U D Y ......................
10
DEFINITION OF TERMS....................................
11
SUMMARY................................................
12
REVIEW OF LITERATURE..............................
WARMTH AND ITS EFFECT UPON LEARNING. . . . . . . . . . .
Summary..........
13
14
35
BEHAVIORAL CUES OF WARMTH..............................
37
Summary..............................................
45
PRESTIGE AND ITS AFFECT UPON LEARNING..........
46
Summary..............................................
64
SUMMARY. '..............................................
65
V
Chapter
3.
Page
PROCEDURES..............
70
COMMUNITY DESCRIPTION.................................. •
70
DESCRIPTION OF THE POPULATION..........................
71
SAMPLING PROCEDURE ....................................
72
TYPES OF DATA COLLECTED................................
73
Teacher Behaviors and Characteristics,................
74
Reliability........................ . . I ..........
79
Validity......................
81
Data Relating to Teacher Warmth......................
81
Reliability....................
81
Validity............................................
82
Data Relating to Teacher Prestige. ..................
83
Reliability. . ......................................
84
Validity............................................
84
Data Relating to Teacher Effectiveness . .............
85
Teacher Self-Eatings ......................
. . . . .
87
METHOD OF COLLECTING DATA. ............
87
HYPOTHESES........................................
88
ANALYSIS OF D A T A ............ .'.......................
90
SUMMARY..........................
91
vi
Chapter
4.
Page
ANALYSIS AND R E S U L T S ......................................
92
ANALYSIS OF D A T A ........................................
93
Null Hypothesis
I...................................
93
Null Hypothesis2. . . . ......................
97
Null Hypothesis
3..................................... 101
Null Hypothesis
4 .........................
105
Null Hypothesis
5...................
107
Null Hypothesis
6.............................
Ill
Null Hypothesis
7..................
112
Null Hypothesis
8.............
113
Null Hypothesis
9..................................... 118
DISCUSSION OF RESULTS.................................... 119
SUMMARY....................................
,5.
SUMMARY, CONCLUSIONS, AND RECOMMENDATIONS............ . .
13
132
SUMMARY. ................................................. 132
CONCLUSIONS. ................................
141
RECOMMENDATIONS.......................................... 144
APPENDICES...................................... •........... 146
A.
Student PerceptionsScale...............................
147
B.
Teacher Behaviors and Characteristics Checklist.......... 150
C.
Letter to Teachers .............................
LITERATURE C I T E D .............................................. .153
152
vii
LIST OF TABLES
Table
Page
1.
Summary of Results of "Verbal Conditioning" Studies . . . .
26
2.
Population Distribution by College.............. .........
72
3.
Least-Square Means and Analysis of Variance Results in
Comparing Expected Grades and Teacher Performance . . . .
94
Significant Results of the Duncan's Test for Expected
Grades Versus Teacher Performance .............. . . . .
95
Least-Square Means and Analysis of Variance Results in
Comparing Teacher Self-Ratings and Teacher Performance^ .
98
4.
5.
6.
Significant Results of the Duncan's Test for Teacher SelfRatings Versus Teacher Performance.......................
99
7.
Least-Square Means and Analysis of Variance Results in
Comparing Teacher Rank and Teacher Performance. ......... 102
8.
Significant Results of the Duncan's Test for Teacher Rank
Versus Teacher Performance.....................
9.
10.
11.
12.
13.
14.
103
Least-Square Means and Analysis of Variance Results in
Comparing Teacher Title and Teacher Performance .........
106
Significant Results of the Duncan's Test for Teacher Title
Versus Teacher Performance. . . . . . .....................
107
Least-Square Means and Analysis of Variance Results in
Comparing Teacher Dress and Teacher Performance .........
108
Significant Results of the Duncan's Test for Teacher Dress
Versus Teacher Performance...............................
109
Correlation Coefficients for Teacher Behaviors and
Characteristics.Versus Teacher Performance, . . . . . . . .
112
Least-Square Means and Analysis of Variance Results in
Comparing Teacher Negative Evaluation and Teacher
Performance.............................................. 114
viii
Table
.15.
16.
17.
Page
Least-Square Means and Analysis of Variance Results in
Comparing Teacher Positive Experiences and Teacher
Performance . . . . . ..................................
115
Significant Results of the Duncan's Test for Teacher
Positive Experiences Versus Teacher Performance ........
116
Correlation Coefficients for Teacher Performance Versus
Teacher Performance.......... .. ............. ..........118
j
ix
ABSTRACT
The intent of this study was to find out if there was any
relationship among selected classroom behaviors and characteristics of
a sample of teachers at Montana State University and student perceptions
of teacher warmth, prestige, and effectiveness. In addition, this
study sought to examine the relationships among these teacher behaviors
and characteristics and teacher self-ratings and the grades the
students expected to receive for the course.
Student volunteers were trained in the use of the Teacher
Behaviors and Characteristics Checklist, an instrument developed by the
investigator, and then attended three classes of each of the teachers
involved in the study during Spring Quarter 1975 in order to count and
categorize teacher behaviors and characteristics. Three instruments
were administered to the students of these classes— the Authoritative­
ness Scale, the Scale for Measurement of Counselor Traits, and the
Revised Faculty Rating Form. Data relating to the grades the students
expected to receive for the course and teacher self-ratings as to the
effectiveness of their instruction was also collected.
The hypotheses tested in this study were concerned with the
relationships among teacher behaviors and characteristics and five
independent measures of teacher performance. These five measures
were teacher warmth, teacher prestige, teacher effectiveness, course
value— item 33, and course value— item 85. All hypotheses were tested
at either the .05 or the .01 level of significance using either the
analysis of variance or the Pearson correlation.
Results of this study relevant to the measure of warmth indicated
that teachers who were perceived by students as being warmer persons
were of higher academic rank, held doctor's degrees, rated themselves
between "average" and "above average" in effectiveness, related more
of their positive experiences to their class, and tended to dress more
in ties and dress pants, sports coats and dress pants, and suits' but
did not wear sports coats and suits all of the time. They also posi­
tively evaluated student responses less frequently, did not acknowledge
student feelings as often, smiled less, and their students spoke less,
in class and expected to receive lower grades for the course.
Teachers who were viewed with more prestige rated themselves
between "average" and "above average" in effectiveness, held doctor's
degrees, were expected to give lower grades, and were of higher academic
rank— with one notable exception. Instructors received the highest
ratings.
X
Teachers who received higher ratings in effectiveness rated
themselves between "average" and "above average" in effectiveness,
had classes in which student laughter occurred more frequently, were
of higher academic rank, related more of their positive experiences to
their class, and dressed less casually,, tending to wear ties and dress
pants, sports coats and dress pants, and suits. They also positively
evaluated student responses less frequently, did not acknowledge
student feelings as often, smiled less, and their students spoke less
in class.
The results of this study also indicated that warm teachers
were viewed with more prestige and were seen as more effective. How­
ever, the courses of warm teachers and prestigious teachers were
valued less by the students.
Chapter I
INTRODUCTION
Adult and higher education is a topic of vital interest in our
nation today.
The day is here when man in a world of increased
technology, increased leisure time, and increased social interaction
is finding that education is indeed a life-long process.
Along with
this increased significance of education is a greater need for more and
better adult educators.
Probably no aspect of education has been discussed with
greater frequency, with as much deep concern, or by more edu­
cators and citizens than has that of teacher effectiveness—
how to define it, how to identify it, how to measure it, how
to evaluate it, and how to detect and remove obstacles to its
achievement . . . .
But findings about the competence of
teachers are inconclusive and piecemeal; and little is presently
known for certain about teacher excellence (Biddle and Ellens,
1964:5).
Studies have been done which go far to answer the question as
to the sort of procedures which have been found.successful in estab­
lishing good relationships in a classroom and thus contribute to
teacher effectiveness.
However, they do not help the teacher to
answer other questions which arise on actually encountering students.
"Have I the correct sort of personality?"
"Will I be able to win and
\
hold their attention?"
It was the intept'of this study to determine and describe some
of the characteristics which constitute the "correct sort of person­
ality" and will enable an instructor to win and hold the students'
2
attention.
Two characteristics which may play an important role in the
teaching-learning process are personal warmth and social prestige.
It
appears that warmth or positive regard from others is a genuine need
of each individual and thus constitutes a powerful social reinforcer.
A teacher who responds to his students in such a way as to contribute
positively to their feelings of self-worth becomes an important person
to those students and capable of modifying their behavior.
The
effectiveness of a teacher’s social reinforcement and also his potency
as a social model are further enhanced if he is held in esteem by his
students.
In other words, not only is it rewarding to an individual
to be around another who expresses positive feelings toward him, it
is even more rewarding if that person is of high prestige.
David G. Ryans (I960) notes that it is of interest to consider
the kinds of behaviors people remember about teachers and to raise the
question of relative importance of such remembered characteristics with
respect to behaviors normally assumed to characterize teaching.
Using
a critical-incidents approach in his teacher characteristics study,
Ryan found that most teaching incidents reported (descriptions of
actual observed behaviors believed to have contributed to the judgment
of superiority or inferiority of the teachers) involved personal or
social teacher behaviors, even though directions had given the judges
complete freedom in naming critical incidents.
He then poses and
3
answers the following question:
Are personal or social characteristics more important than
a teacher's scholarliness, the teaching procedures followed,
unique demonstrations, or the content taught? One may well doubt
they are more important, but they may be equally important. We
question why more people often do not mention incidents
involving the teaching learning process per se . . . all of us
seek personal reinforcement and it is in the area of the personal
or social characteristics of teachers and other persons that we
best recall events (Ryans, 1969:72).
It is the contention of the writer that the more a teacher is
held in esteem by his students and the more he contributes verbally
and nonverbally to their feelings of self-worth the more effective he
will be in influencing their behavior.
A teacher is thus in a position
to reinforce learning in his students in either an appropriate or
inappropriate way, either wittingly or unwittingly.
For him to be an
effective teacher he should be aware of those verbal and nonverbal
behaviors which are indicators of warmth and acceptance to his students,
and also those characteristics and behaviors which might cause him to
be held in esteem by his students.
Asche (1946:258) points out:
We look at a person and immediately a certain impression of
his character forms itself in us. A glance, a few spoken words
are sufficient to tell us a story about a highly complex matter.
We know that such impressions form with remarkable rapidity and
with great ease. Subsequent observation may enrich or upset our
first view, but we can no more prevent its rapid growth than we
can avoid perceiving a given visual object or hearing a melody.
We also know that this process, though often imperfect, is also
at times extraordinarily sensitive.
It is for these reasons that the characteristics of teacher
4
warmth and prestige are important aspects of the teacher-learning
process and merit investigation.
STATEMENT OF THE PROBLEM
The problem of this study was to find out if there was
any relationship among selected classroom behaviors and character­
istics of a sample of teachers at Montana State University and
student perceptions of teacher warmth, prestige, and effectiveness.
In addition, this study sought to examine the relationships
among these teacher behaviors and.characteristics and teacher selfratings and the grades the students expected to receive for the
course.
PURPOSE OF THE STUDY
i
The purpose of this study was to determine some of those
/
interpersonal characteristics and behaviors that make up the person­
ality of a good teacher.
In other words, it was concerned with the
question, "What personal qualities make a teacher effective?"
This
research focused on those teacher characteristics and behaviors which
seem to be related to the degree to which teachers are held in esteem
by their students and the extent to which they are viewed as warm
5
individuals by their students.
.Xx
Furthermore, it was the purpose of this study to define the
characteristics of warmth and prestige in behavioral terms in order
that an individual so desiring might endeavor to incorporate them into
his personality.
NEED FOR THE STUDY
A survey of the available literature pertinent to the issues
of this study indicated that, although studies have been done which
attempted to determine the interpersonal characteristics of a good
leader, counselor, or educator, these characteristics, for the most
part, were not behaviorally defined and/or were not examined in
relation to adult learning.
Much money, time, and effort is involved in the development by
our educational system of teachers— hopefully good teachers^--in order
to provide the best possible educational opportunities for the country's
citizens.
This task is undertaken without a complete knowledge of what
constitutes a good teacher.
Teachers themselves do not know what makes
them effective and, for the most part, rely on "seat of the pants"
technology.
The lack of an adequate, concrete, obi active, universal
criterion for teaching ability is thus the primary source of
trouble of all who would measure teaching. One typical method
of attack is to compile a list of broad general traits supposedly
desirable for teachers, with respect to which the rater passes
judgment on each teacher. This amounts to an arbitrary
6
definition of good teaching, which is subjective and usually
vague, but it does not necessarily lead to. a definition of it.
Only if the traits themselves can be reliably identified can
their possessor be identified as a "good teacher" according to
the definition laid down in the scale (Lancelot, Barr, Torgeson,
Johnson, Lyon, Walvoord, and Betts, 1935
After reviewing over 150 articles on the personality charac­
teristics of teachers, Getzels and Jackson (1963:574) concluded that:
Despite the critical importance of the problem and halfcentury of prodigious effort, very little is known for certain
about the nature and measurement of teacher personality, or about
the relation between teacher personality and teacher effective­
ness. The regrettable fact is that many of the studies so far
have not produced significant results.
Another discouraging aspect of the attempt to ascertain what
makes a teacher effective is the fact that learning in an educational
setting is conditioned by a very large number of variables.
These may
include such governing factors as intelligence, the pupils' own habits
of study, interest, and physical condition.
Learning may also be
affected by factors associated with the teacher, his personality,
voice, dress, clarity of thought and expression, sense of humor, and
so on.
It is these latter kinds of factors which serve as the focus
of this study.
Many studies have examined the flow of verbal interaction
between teachers and their students, i.e., the Flanders system.
In
such studies verbal behaviors are categorized and classrooms are
observed in order to encode student and teacher verbal behaviors.
addition to the charting of verbal interactions, attention must be
In
7
paid to the nonverbal behaviors of both the teacher and his students
in order to obtain a picture of the total interaction process in the
classroom.
Hendrix (1960:39) has observed that:
One phase of teaching looming large in things revealed to
date is the enormous role played by nonverbal communication
between teacher and students. Current research in paralinguis­
tics— especially that involving kinesics— is revealing ways to
identify and classify the nonverbal behavior which human beings
learn to interpret in each other; it is thus that we produce
the complicated stream of communication sometimes accompanying,
sometimes independent of words. Such an analysis might enable a
teacher to cultivate desirable paralinguistic effects, and to
avoid those which are destructive to his work.
There is a need to define teacher interpersonal characteristics
in behavioral terms such that a greater insight into the teaching­
learning process might be provided.
Also, characteristics so described
might be more easily taught, more easily learned, and more amendable
to experimental analysis.
X
GENERAL QUESTIONS TO BE INVESTIGATED
It is the intent of this study to obtain information bearing
on the following questions.
I.
Are there identifiable and quantifiable characteristics and
behaviors of teachers which would affect the degree to which they are
viewed as warm, and the degree to which they are viewed with prestige?
'2.
Do observer ratings of teacher behaviors and characteristics
differentiate between teachers who are warm and those who are not.
8
between teachers who are viewed with prestige and those who are not and
between effective and non-effective teachers?
3.
Do student perceptions of teacher warmth differentiate
between effective and non-effective teachers?
4.
Do student perceptions of teacher prestige differentiate
between effective and non-effective teachers?
GENERAL PROCEDURES
The problem was approached by first determining which instru­
ments were available that would be capable of assessing the character­
istics, behaviors, and perceptions needed to answer the general
questions.
The investigator decided to use the Authoritativeness
Scale (McCroskey, 1966), the Scale for Measurement of Counselor Traits
(Suvak, 1966), the Revised Faculty Rating Form (Miller and Guinouard,
1966), and the Teacher Behaviors and Characteristics Checklist.
The
last instrument was developed by the investigator.
The next step was to select, and obtain the cooperation of, the
teachers who participated in the study.
This was accomplished by
obtaining a random sample of all.the 300 level courses offered during
Spring Quarter, 1975, an<I then asking the teachers of these courses to
participate in the study.
Finally, student volunteers from three sections of Educational
Psychology 208 were trained in the use of the Teacher Behavior and
9
Characteristics Checklist and then attended three classes of each of
the teachers during the quarter in order to count and categorize
teacher behaviors and characteristics.
The data was collected during the Spring Quarter, 1975,
organized, and analyzed following the collection.
This paper has been structured along the following outline.
In Chapter I, an introduction to the problem was presented, a state­
ment of the problem was given, the need and purpose of the study was
clarified, and general questions to be investigated were considered. .
General procedures were described, limitations acknowledged, and a
definition of terms were given.
Chapter 2 consisted of a review of selected literature deemed
pertinent to the problem and questions presented in Chapter I.
Litera­
ture reviewed included that of teacher warmth and its affect upon
learning, behavioral cues of warmth, and teacher prestige and its
affect upon learning.
Chapter 3 consisted of the research procedures.
Included were
the description of the community and population, methods of collecting
the data, the reliability and validity of the instruments used, and the
organization and analysis of the data.
Chapter 4 consisted of the analysis and the results of the
data collected.
I'
I
10
LIMITATIONS OF THE STUDY
1.
This study was limited to Montana State University students
enrolled in 300 level courses during Spring Quarter, 1975.
2.
The fact that some teachers were unwilling to participate
in the study may have been a limiting factor.
There is the possibility
that a certain type of teacher refuses to cooperate with such a study.
3.
Another limitation of this study was the possibility that
teacher-student contact outside the classroom colored student percep­
tions of their teacher.
4.
The general student and teacher population at Montana State
University may have been a limiting factor.
Due to the nature of its
curriculum offerings and geographical setting, a certain type of
student and/or teacher, not representative of students and/or teachers
nationwide, may be attracted to Montana State University.
5.
The resources available at the Library of Montana State
University may have been a limiting factor.
Financial resources
available and selection recommended by the faculty limit the scope and
variety of resources.
' 6.
The selection of the literature to be reviewed by the
investigator may have been a limiting factor.
The investigator
restricted his review to certain selections and may not have reviewed
other appropriate literature.
7.
The instruments used to collect the data, the
11
Authoritativeness Scale, the Scale for Measurement of Counselor Traits,
the Revised Faculty Rating Form, and the Teacher Behaviors and Charac­
teristics Checklist may have been limiting factors.
Other instruments,
unknown to the investigator, might have been more appropriate for the
study.
DEFINITION OF TERMS
The following is a list■of terms and their definitions as they
were used throughout the study.
Behavior.
For the purposes of this study, behavior is definied
as anything a person does which is observable or recordable, i.e.,
words, mannerisms, etc.
Characteristic.
The term characteristic is defined as those
aspects of a person which are not behaviors, i.e., dress, title, etc.
Student Perceptions Questionnaire.
This questionnaire consists
of the Authoritativeness Scale, the Scale for Measurement of Counselor
Traits, and the Revised Faculty Rating Form.
Teacher Behavior and Characteristics Checklist.
The Teacher
Behavior and Characteristics Checklist was an instrument developed by
the investigator to categorize and quantify teacher behaviors in the
classroom.
This scale consists of eleven categories which enable an
12
observer to Identify and count the frequency of such teacher behaviors
as expressing personal concerns, smiling, and attending behaviors.
SUMMARY•
A need for investigation in the area of teacher personality
and its relation to teacher effectiveness is apparent.
Although
N1
studies have been done which sought to determine the interpersonal
characteristics of a good leader, counselor, or educator, these
characteristics, for the most part, were not behaviorally defined
and/or were not examined in relation to adult learning.
The investigator viewed this study as an attempt to identify
and behaviorally describe those interpersonal characteristics'' and
behaviors that make up the personality of a good teacher.
The study
had as its focus those teacher characteristics and behaviors which seem
to be related to the degree to which teachers are held ^in esteem by
I
their students and the extent to which they are viewed as warm
individuals by their students.
This chapter presented an introduction to the concepts of
teacher warmth and prestige and posed several questions relating to the
possible effects they may have upon teacher effectiveness.
The need
for the study, general procedures for conducting the study, limitations,
and definitions of terms were presented.
Chapter 2 includes a review of the literature relating to this
particular study.
Chapter 2
,
REVIEW OF LITERATURE
The primary emphasis of this chapter is the review of literature
relating to the personal characteristics of warmth and prestige and the
role they play in interpersonal communication, especially as they
pertain to the teaching-learning process.
In addition, this chapter
will examine the specific behaviors that convey warmth, that is, those
behaviors which tell another that he is regarded in a positive way.
The chapter is then organized around the following three themes.
1.
Warmth and its affect upon learning.
This section is a
review of the literature concerned with the characteristics of warmth
and the role it plays in interpersonal relationships, especially as it
relates to the teacher-student relationship.
Teacher warmth is viewed
from the framework of learning theory as a powerful reinforcer of
student learning.
Studies investigating the effectiveness of certain
verbal and nonverbal responses of an experimenter in influencing a
subject's behavior and in some cases nonverbal behavior are reported.
2.
Behavioral cues of warmth.
It is the aim of this section
to present those studies which attempt to define and specify the
behavioral cues of warmth.
3.
Prestige and its affect upon learning.
This section deals
largely with studies of social influence which look at communicator
prestige as a factor in how effectively a message is communicated.
14
That is, to what degree and in what way does the prestige of the
communicator affect his ability to modify the behavior of others.
Studies relating the prestige of a source to his potency as a social
model are also discussed.
WAEMTH AND ITS AFFECT UPON LEARNING
If psychotherapy can be looked upon in a general way as a
learning process, then it is relevant to this study to consider the
theory of personality of Carl R. Rogers (1952:483) which was evolved
from the study of adults in therapy.
One of Rogers' basic propositions is that the more an individual
is able to attend to, think about, and accept as part of himself the
whole range of his responses, the better adjusted he is likely to be,
and that his most important self-conceptions are learned through his
interactions with other people (Ford and Urban, 1963;410).
Ford and Urban (1963:410) go on to describe Rogers' ideas of
"need for positive regard and self-regard":
... . another important product of learning involves habits
of seeking certain kinds of consequences and affective responses
related to them. Here the emphasis is on interpersonal situa­
tions and the behavior of other people toward the individual.
As the infant becomes aware of himself as an entity different
from others, he begins to notice differences in their responses
to him and his to them. One important category of such responses
Rogers calls "positive regard." This includes such responses as
"warmth, liking, respect, sympathy, acceptance," all of which
seem to have a common denominator of positive affect', an appar­
ently innately desired response which one seeks to create in
1
>
15
oneself.
It Is proposed that when the individual notices that
others are responding toward him with positive affect, it elicits
positive affect in him— it is satisfying.
To put it differently, when others evaluate a child's responses
negatively by displaying anger or disapproval, discomforting
affective responses are produced in the child. Positive evalua­
tion by others through smiles, approvals, or affective responses,
however, produce "satisfying" affect in the child. The child
gradually comes to seek the latter and avoid the former. In
Rogers' terms, he acquires a need for positive regard.
Harris (1973:68) also contends that individuals have a need
for positive regard from others.
This positive evaluation or recogni­
tion from others he terms stroking.
In the "I'm not OK— You're OK"
position, which is the universal position of early childhood— a position
most of us maintain at least in part the rest of our lives— the person
feels at the mercy of others.
"He feels a great need for stroking, or
recognition, which is the psychological version of the early physical
stroking."
That warmth is an important quality in a person which seems
to carry more weight than others in establishing a view of an individ­
ual's personality was investigated by Asch (1946:258).
The basic plan
followed in the series of experiments he reported was to read to the
subject a number of discrete characteristics, said to belong to a
person, with the instruction to describe the impression he formed.
It was found that the characteristic "warm-cold" produced striking
and consistent differences of impression with the characteristic warmth
producing a far more positive impression than the characteristic cold.
\
16
Whereas the warm-cold variable had been found by Asch (1946)
to produce large differences in the impressions of personality formed
from a list of adjectives, Kelly (1950:431) introduced the same vari­
able in the form of expectations about a real person in a classroom
setting, and obtained similar results.
Before his actual appearance
in a classroom as a substitute instructor, the stimulus person was
introduced by an experimenter, and a little biographical note about
him was passed out to the students randomly in such a manner that they
were not aware that two kinds of information was being given out.
The
two notes were identical, except that in one the stimulus person was
described among other things as being "rather cold" whereas in the
other form the phrase "very warm" was substituted.
It was found that
different first impressions were produced by the different expectations; ^
and they were shown to influence the observers' behavior toward the
stimulus person.
Those observers given the favorable expectation (who,
consequently, had a favorable impression of the stimulus person)
tended to interact more with him than those given the unfavorable
expectation.
In summarizing the statistically significant differences in
the way subjects viewed the stimulus person and rated him on a set of
fifteen rating scales, Kelly (1950:431) states:
The "warm" subjects rated the stimulus person as more
considerate of others, more informal, more sociable, more popular,
better natured, more humorous, and more humane. These findings
are very similar to Asch's for the characteristics common to both
17
studies. He found more frequent attributions to his hypothetical
"warm" personalities of sociability, popularity, good naturedness,
generosity, humorousness, and humaneness. So these data strongly ■
support his finding that such a central quality as "warmth" can
greatly influence the total impression of a personality. This .
effect is found to be operative in the perceptions of real
persons.
A study by Lehat-Mandelbaum and Kipnis (1973:250) indicates
that teacher consideration is an important factor in the teaching­
learning process.
They asked college students to describe the behavior
of their instructors using an adaptation of Fleishman’s Supervisory
Behavior Description Questionnaire and to also evaluate their ability
to teach.
Teacher behaviors were categorized in terms of teacher
consideration, which relates to the teacher’s personal relationship with
students— his attention to emotional and social aspects of students’
classroom life— and in terms of teacher initiating structure which
refers to an emphasis on the content of the course and the learning
tasks.
The authors concluded that "the teacher seen as high in con­
sideration by his students was considered to be the superior teacher."
In a study by Dawson, Messe, and Phillips (1972:369), an
experimenter manipulated his consideration and initiating structure
behaviors while teaching four sections of general psychology, and
found that classes taught with high consideration were higher on three
dependent measures of performance than students taught with low
consideration.
Students in classes taught with high consideration
performed higher on the submission of annotated bibliographies,
18
answering test items correctly, and obtaining research credits.
In
classes taught with high initiating structure, students performed
higher on the submission of annotated bibliographies.
Gage (1972:35) examined research using such process measures
as the Minnesota Teacher Attitude Inventory and the Flanders' inter­
action categories and found the following:
(a.) Teachers differ reliably from one another on a series
of measuring instruments that seem to have a great deal in common,
(b.) These reliable individual differences among teachers are
fairly consistently related to various desirable things about
teachers...Teachers at the desirable end tend to behave approv­
ingly, acceptantly, and supportively; they tend to speak well of
their own students, students in general, and people in general.
They tend to like and trust rather than fear other people of all
kinds.
Thus, there is evidence which supports the thesis that the
characteristic of warmth is important in interpersonal relationships
and may be an important factor in determining a teacher’s effective­
ness.
In the teacher-student relationship, teacher warmth and positive
regard might be viewed from the framework of learning theory as
powerful reinforcers of student learning.
There are a number of
reports in the literature, generally appearing under a verbal condi­
tioning label, which indicate that teacher or experimenter verbal and
non-verbal behaviors which might be construed as signs of approval or
regard are effective reinforcers.
These studies are an application of
learning theory in that they deal with the operant conditioning of
verbal behavior.
19
B. F. Skinner (1957:53) pointed out that:
The effect of this procedure in releasing a response from a
specific controlling condition is usually achieved in another
way. Instead of using a great variety of reinforcements, each
of which is relevant to a given state of deprivation or aversive
stimulation, a contingency is arranged between a verbal response
and a "generalized conditioned reinforcer." Any event which
characteristically precedes many different reinforcers can be
used as a reinforcer to bring behavior under the control of all
appropriate conditions.of deprivation and aversive stimulation.
A response which is characteristically followed by such a
generalized conditioned reinforcer has dynamic properties similar
to those which it would have acquired if it had been followed by
all the.specific reinforcers of issue.
A common generalized conditioned reinforcer is "approval."
It is often difficult to specify its physical dimensions. It
may be little more than a nod or a smile on the part of someone
who characteristically supplies a variety of reinforcements.
Sometimes . . . it has a verbal form: "Right!" or "Good!"
Because these signs of approval frequently precede specific
reinforcements appropriate to many states of deprivation, the
behavior they reinforce is likely to be in strength much of the
time.
The effectiveness of certain verbal and nonverbal responses of
an experimenter in conditioning a subject's verbal behavior and in some
cases nonverbal behavior has been demonstrated in many studies.
Weiss,
Krasner, and Ullmann (1960:415) experimentally manipulated the
emotional atmosphere of examiner reinforcement to determine its effect
on interpersonal responsiveness as measured by changes in samples of
complex verbal behavior.
During reinforced trials, the experimenter
verbally reinforced the subject's use of emotional words by saying
"mmm-hmm" and nodding his head as if in agreement.
The emotional
atmospheres were induced by the experimenter during two distinctly
different.interactions with the subject, each lasting approximately ten
20
minutes.
Half the subjects had been assigned at random to the
hostility condition in which they were interviewed by an experimenter
who was openly critical, impatient, non-believing, and who frequently
expressed his disappointment of the subject's answers, and by implica­
tion, questioned the subject's suitability for college work.
For
comparison the other half of the subjects were exposed to a mildly
positive interaction with the same experimenter during which the
conversation centered on the subject's interests, aims, and academic
work.
"Throughout, the experimenter maintained a friendly relaxed
atmosphere by expressing interest in the subject."
It was found that
the induction of a hostile atmosphere significantly reduced responsive­
ness to conditioning.
Sapolsky (1960:241) provided support for the hypothesis that
social reinforcement from a high attraction source is more effective
than from a low attraction source in two studies which manipulated high
or low attraction between subjects and the experimenter.
The subjects
in the high attraction group were told, "You will be paired with an
experimenter whom you will find congenial.
tionnaire which enables us to do this."
We have developed a ques­
Subjects in the low-attraction
group were told, "Usually, we can match people quite well, but in your
cases we're going to have some trouble.
It's going to take too long to
locate someone for you, so I'm assigning you to Miss C.
irritate you a little, but do the best you can."
She may
The effectiveness of
21
the directions in establishing the two experimental groups was
verified after the experimental session by the subject's ratings of
the experimenter on a self-anchored sociometric scale.
It was found
that when there was high attraction between the subject and the
experimenter, the reinforcing value of "mmm-hmm" was enhanced as
evidenced by a significant increase in the use of first person
pronouns which made up the reinforced response class.
When the
attraction between the subject and the experimenter was low, no
increase resulted.
When the experimenter left the room and subjects
continued construction of tape-recorded sentences, subjects in the
low attraction condition exhibited a significant increase in the
previously rewarded behavior.
It was concluded that the effect of a
non-attractive or incompatible experimenter was to suppress or counter­
act the immediate effectiveness of the positive reinforcement.
Ferguson and Buss (1960:324) investigated operant conditioning
of hostile verbs in relation to aggressiveness of the experimenter.
Subjects who were instructed to make up a sentence using one verb and
one. pronoun were verbally reinforced for using hostile verbs.
Each of
the two experimenters played a neutral role with one group and an
aggressive role with the other.
"The neutral experimenter was patient,
calm, and courteous in an attempt to establish a non-hostile experimen­
tal climate.
The aggressive experimenter was brusque, unfriendly,
impatient, and tended to scowl and sneer."
Reports from the subjects
22
at the termination of the experiment revealed that the two roles
(neutral or aggressive) were enacted appropriately.
It was found that
the aggressiveness of the experimenter significantly affected condi­
tioning:
an aggressive experimenter retards learning in comparison
to a neutral experimenter.
Employing a verbal conditioning paradigm with verbs of a
"mildly hostile" connotation as the reinforced response class, Sarason
(1962:376) found that a greater learning effect was obtained from high
hostile subjects run by low hostile experimenters.
The degree of
subject and experimenter hostility was determined by their scores on
the Hostility Scale of Sarason1s Autobiographical Survey (Sarason,
1958).
Sarason pointed out that, "One possibility suggested by these .
results is that the level of the subjects responsiveness may be influ­
enced by the degree of experimenter-subject compatibility or similarity."
The results of a later study by Sarason and Minard (1963:87) indicated
that subjects run by high hostile experimenters exhibited a learning
effect only in the high prestige condition, that is, when the experi­
menter was viewed as prestigious by the subject.
In an actual classroom situation, the reinforcement of student
responses may take other forms than a teacher simply saying "mmm-hmm,"
"yeah," or "good" each time the student responds in a way the teacher
considers appropriate.
and Simon (1969:68)
Knowles (1973:90) reviewed a study by Flanders
who concluded from their examination of a dozen
23
studies that:
The percentage of teacher statements that make use of ideas
and opinions previously expressed by pupils is directly related
to average class scores on attitude scales of teacher attractive­
ness, liking the class, etc. , as well as to average achievement
scores adjusted for initial ability.
Matarazzo, Saslow, Wiens, Weitman, and Allen (1964:54)
studied the effect of interviewer headnodding on interviewee speech
behavior.
In this experiment, a control period of no interviewer
headnodding was followed by a period in which the experimental variable
was introduced; i.e., each time the interviewee began an utterance,
the interviewer nodded his head repeatedly throughout that whole
utterance.
The results showed that the period of interviewer head­
nodding was associated with an increase in the average interviewee
speech duration.
Cientat (1959:648) reported the effects of nonverbal gestural
cues on rate of verbalization in a free-responding conversational
situation; the attention to student responses by the professor and a
student confederate was varied.
Attention or positive reinforcement
consisted of students being looked at, whenever they spoke, and being
given occasional nods of approbation.
Ignorance or negative reinforce­
ment was shown by the professor and a confederate looking away from
students when, the latter spoke.
Cientat found that the amount of time
during which a subject spoke was a "positive function of attention and
a negative function of inattention."
24
In'another study employing a verbal conditioning paradigm,
Matarazzo, Wiens, Saslow, Allen, and Weitman (1964:109), an inter­
viewer's "mmm-hmm" was used as the verbal social reinforcing stimulus.
A control period during which the interviewer did not say "mmm-hmm"
was followed by a period in which he said "mmm-hmm" throughout each of
the subject's utterances.
Results indicated that the average inter­
viewee speech duration was greater during the period in which the
interviewer.said "mmm-hmm" than during the period in which he did not.
Two studies by Weiss, Krasner, and Ullmann (1963:423) and Ikman,
Krasner, and Ullmann (1963:387) replicated the results of these verbal
conditioning procedures employing experimenter headnodding and his
saying "mmm-hmm" as social reinforcers.
In a study by Krasner, Ullmann,
Weiss, and Collins (1961:411) which also demonstrated the verbal condi­
tioning phenomenon, it was noted the two male experimenters obtained
significantly greater use of the specified verbal class, emotional
words, during reinforced trials rather than operant trials, while the
female experimenter obtained group means in the same direction as the
two male experimenters, but not to a statistically significant extent.
Although the authors do not discuss the point, it is possible that the
subjects were influenced to a greater extent by the male experimenters
who were Ph.D's in psychology and introduced themselves as "Doctor"
than by the female experimenter who had an A.B. in psychology and
introduced herself as "Miss," because they viewed the male experimenter
25
as more prestigious.
This point is discussed extensively in the next
section of this chapter.
There are many studies of operant conditioning of verbal
behavior which offer supportive evidence for the notion that experi­
menter behaviors which convey acceptance or positive regard, indeed,
"signs of approval" as Skinner put it, are effective reinforcers.
For
example, a study by Krasner (1958:148) reported on thirty studies, all
of which follow a Skinnerian paradigm in that the dependent variables
are the subject's verbal behavior and the independent variables are
generalized conditioned reinforcers intended to bring verbal behavior
under, the control of the examiner. He reviewed these studies in terms
of the different experimenter verbal and nonverbal behaviors employed
as reinforcing stimuli.
He found the most widely used examiner verbal
behavior to be the emission of the "mmm-hmm" sound, while other verbal
cues in these studies included "good," "uhha," "yeah," "I see," "that's
accurate," "that's right on the button," "that's a good one," "give
another please," "you're right," "right," "all right," "fine," "I
agree," paraphrase of subject's response, and repetition of subject's
response.
The gestural cues included headnodding, headshaking, and
smiling.'
A summary of the results of the studies reviewed by Krasner
is presented in Table I, pages 26 and 27, which is similar to the table
found in his report.
These studies indicate the effectiveness of social approval as
26
Table I
Summary of Results of "Verbal Conditioning" Studies
Author
Reinforcing Stimuli
Class of Behavior
Reinforced
Positive Results
Ball (1952)
"mmm-hmm"
"animal"
Greenspoon (1954)
"mmm-hmm"
plural nouns
Handler & Kaplan (1956)
"mmm-hmm"
plural nouns
B.Sarason (1957)
"mmm-hmm"
verbs
I.Sarason (1957)
"mmm-hmm"
"verbal activity" verbs
Mock (1957)
"mmm-hmm," headnod
"mother"
Krasner (1958)
"mmm-hmm," headnod,
smile
"mother"
Salzinger & Pisoni
(1957) (1957)
"mmm-hmm," "uh-ha,"
or I see
affect statements
Wilson & Verplank (1956) "mmm-hmm," "good,"
or writing
plural nouns, adverbs,
or travel verbs
Binder, et al (1957)
"good"
"hostile" verbs
Cohen, et al (1954)
"good"
"I," "we" pronouns
Cushing (1957)
"good"
"like" person in pictures
Grossburg (1956)
"good"
"I," "we" pronouns
Ekman (1958)
( "good"
anti-capital punishment
response
Hartman (1953)
"good"
"I," "we" pronouns
Hildum & Brown (1956)
"good"
"attitudes"
Klein (1954)
"good"
"I," "we" pronouns
Nuthmann (1957)
"good"
"acceptance of self"
Taffel (.1955)
"good"
"I," "we" pronouns
Tatz (.1956)
"good"
a pair of digits
Fahmy (1953)
"good-one"
human responses
Spivak & Papajohn(1957)
"right"
autokinetic effect
27
Table I (continued)
Author
Reinforcing Stimuli
Class of Behavior
Reinforced
Positive Results
Wickes (1956)
"fine," "good," or
"all right"
movement responses
Wickes (1956)
headnod, smile, or
lean forward
movement responses
Ekman (1958)
headnod, smile, or
lean forward
.movement responses
Verplank (1955)
paraphrase, agree­
ment, smile
opinions
Kanfer (1954)
"that's accurate,"
etc.
autokinetic effort
Hartman (1955)
head shake^
"I," "we" pronouns
Mock (1957)
head shake^
"mother"
Greenspoon (1955)
"huh-uh"b
plural nouns
Daily (.1953)
"mmm-hmm"
"I," "we" pronouns
Hildum & Brown (1956)
"mmm-hmm"
"attitudes"
Cushing (1957)
"good"
"dislike" persons in
pictures
Daily (1953)
"good"
"I," "we" pronouns
Marion (1956)
"good"
"I," "we" pronouns
Hartman (1955)
headnod
"I," "we" pronouns
Fahmy (1953)
repetition of
response
human responses
Fahmy (1953)
"give another one,
please"
human responses
'
28
a conditioned secondary reinforcer.
The question arises as to whether
an experimenter can behave too warmly and thus lose his effectiveness
as a reinforcer.
Salzinger (1959:66) recognizes this possibility when he states:
While some reinforcement theorists have made an attempt to
define a reinforcement independently of its effect upon behavior,
such efforts are largely ineffective when applied to secondary
reinforcements. With a primary reinforcement like food it is
possible to predict its effectiveness from the operation of food
deprivation. It is more difficult to find similar operations for
secondary reinforcements like the utterance "mmm-hmm," a smile,
or a nod of the head.
Notwithstanding this, Gerwitz and Baer (1958:49) made an
attempt to use a social deprivation operation.
Employing children as
subjects they found that when an adult made words and phrases like
"Good!" and "Hm'himn" contingent upon an arbitrarily chosen response,
that response was reinforced (i.e., conditioned).
It was found, in
addition, that this reinforcing effect of approval could be increased
\
when the children experienced
a
preceding twenty minute period of
social isolation, relative to its effectiveness for the same children
when they had not been isolated.
In a later study, Gerwith and Baer
(1958:165) found that behaviors maintained by social reinforcers are
responsive also to a condition of relative satiation for such reinfor­
cers.
Results indicated that the reinforcing effectiveness of approval
was relatively greatest after Deprivation— a period of social isolationintermediate after Non-deprivation, and least after Satiation— equated
to a condition in which an abundance of approval and social contact is
29
supplied to a child by an adult.
It was concluded that:
. . . a reinforcer appearing to be typical of those in
children's social drives appears responsive to deprivation and
satiation operations of a similar order as those controlling
the effectiveness of reinforcers of a number of the primary
drives.
In a study by Simkins (1961:380), subjects were given a series
of tests and were subsequently criticized or complimented on their
performance depending on the treatment condition to which they were
assigned and irregardless of their actual performance.
After the fake
testing, the subjects were presented with a conditioning task, the
learning of hostile verbs.
During this time and until the completion
of the experiment, the experimenter assumed a neutral attitude with
all subjects.
The results indicated that the effectiveness of the
social reinforcers "good" and "that's fine" were dependent upon the
attitude assumed by the experimenter.
Subjects who had been responded
to in an over-solicitous manner by the experimenter tended to show
resistance in being conditioned to use hostile verbs.
Simkins concludes:
Perhaps in the context of extreme social approval a satiation
effect occurs so that the usual social reinforcers lose their
reinforcing effectiveness . . . .
The best situation for the
learning of hostile materials seems to follow a condition of
social disapproval.
It is possible that a teacher might increase his effectiveness
by varying his behavior in such a way that at times he expresses
warmth and acceptance while at other times he appears aloof or even
30
angry.
Johnson (1971:571) states:
If the invariant expression of warmth produces interpersonal
attraction but does not tend to produce influence, the expression
of combinations of warmth and other emotions such as anger may
be more successful in inducing cooperative behaviors from the
listener. A learning theorist who believes in the inhibition of
undesirable behaviors through the use of punishment might suggest
that the expression of anger towards, undesirable behaviors and
. the expression of warmth towards desired behaviors may be far
more effective in inducing cooperation than the invariant expres­
sion of warmth.
Johnson (1971) designed a study, which compared the effectiveness
of expressing different orders of warmth and/or anger upon the induction
of cooperation in the actor and the listener in a negotiating situation.
In the listener part of the study, seven female confederates were given
four hours of instruction .on how to express warmth and anger.
In a
negotiating situation with one subject, the confederate's role was to
express warmth and/or anger.
With subjects in the invariant warmth and
invariant anger conditions, the confederate’s affective expression was
consistent for the entire thirty minutes.
In the warmth-anger and
anger-warmth conditions, the confederate would express one emotion for
the first, half of the negotiating period and the other emotion during
the second half.
In this study, behavioral compliance was measured by
whether the subject would publicly state acceptance of the confeder­
ate's arguments or would publicly statd rejection of his own arguments.
The data indicated that more behavioral compliance took place in the
warmth-ahger and the apger-warmth conditions than in the invariant
31
warmth and the Invariant anger conditions.
Attitude change concerning
the relative merits of the two positions represented was measured by
asking the subjects to indicate the extent to which they felt their
position was superior to the confederate's.
It was found that subjects
in the invariant anger and the warmth-anger conditions felt that their
positions were more superior than did the subjects in the invariant
warmth and the anger-warmth conditions.
The author concludes:
Thus, if one wishes to induce behavioral compliance with one's
position, the expression of warmth followed by anger or the
expression.of anger followed by warmth is more effective than is
the expression of invariant warmth or anger . . . .
But if one
also wishes to change the other's private attitudes concerning
the relative merits of the two positions, the expression of
invariant warmth and the expression of anger followed by warmth
is more effective than the expression of warmth followed by
anger or invariant anger (1971:575).
Aronson and Linder (1965) proposed that the expression of
positive feelings toward another is rewarding to that person, however,
the expression of initially negative feelings, followed by the expres­
sion of increasingly positive feelings might be even more rewarding.
A statement of their viewpoint follows.
It is conceivable that the sequence of 0's behavior toward P
might have more impact on P's liking for 0 than the total number
of rewarding acts emitted by 0 toward P. Stated briefly, it is
our contention that the feeling of gain or loss is extremely
important— specifically, that a gain in esteem is a more potent
reward than invariant esteem, and similarly, the loss of esteem
is a more potent "punishment" than invariant negative esteem.
Thus, if 0's behavior toward P was initially negative but
gradually became more positive, P would like 0 more than he
I,.
32
would had 0's behavior been uniformly positive. This would
follow even if, in the second case, the sum total of rewarding
acts emitted by 0 was less than in the first case (1965:156).
Aronson and Linder tested their hypothesis in the following
manner:
In a laboratory experiment, coeds interacted in two person
groups over a series of seven brief meetings.
After each meeting, the
subjects were allowed to eavesdrop on a conversation between the
experimenter and her partner in which the latter (actually a confeder­
ate) evaluated the subject.
These evaluations involved the confeder­
ate's expression of either a uniformly positive attitude toward the
subject, a uniformly negative attitude toward the subject, a negative
attitude which gradually became positive, or a positive attitude which
gradually became negative.
It is important to note that the positive
evaluations in the Positive-Positive condition were qualitatively the
same as the final two evaluations in the.Negative-Positive condition.
However, there was a quantitative difference.
Because the number of
evaluations was the same in both conditions, the subjects who received
only positive evaluations received a greater number of positive rein­
forcements and fewer negative reinforcements than subjects in the
Negative-Positive condition.
The Positive-Negative condition was the
mirror image of the Negative-Positive condition.
In the Positive-
Negative condition, the confederate began by stating that the subject
seemed interesting, intelligent, and likeable, but by the seventh
session she described the subject as being dull, ordinary, etc.
In
33
the Negative-Positive condition, the confederate began by describing
the subject as dull, ordinary, not very intelligent, but during the
fourth session began to change her opinion about her.
Her attitude
became more favorable with each successive meeting until, in the
seventh interview, it was entirely positive.. The major results showed
that the subject liked the confederate best when her evaluations moved
from negative to positive and least when her evaluations moved from
positive to negative.
It was concluded that a gain in esteem is more
rewarding than continuous positive esteem, and that a loss in esteem
is more punishing than constant negative esteem.
One of the explanations of this gain-loss effect advanced by
the authors is a cognitive one.
By changing his opinion about P , 0 forces P to take his evalua­
tion more seriously. If 0 expresses uniformly positive or
uniformly negative feelings about P , P can dismiss this behavior
as being a function of 0's style of response, i.e., that 0 likes
everybody or dislikes everybody, and that it is his problem. But
if 0 begins by evaluating P negatively and then becomes more
positive, P must consider the possibility that 0's evaluations
are a function of 0's perception of him and not merely a style of
responding. Because of this he is more apt to be impressed by
0 than if 0's evaluation had been invariably positive. It is
probably not very meaningful to be liked by a person with no
discernment or discrimination. 0's early negative evaluation
proves that he has discernment and that he's paying attention
to P— that he's, neither blind nor bland. This renders his
subsequent positive evaluation all the more meaningful and
valuable (1965:168).
Sigall and Aronson (1967) extended the findings of Aronson
and Linder (1965) into a different area— the area of communication and
opinion change.
In this experiment a communicator, prior to presenting
34
his persuasive communication, needed to interact with the subject
over a period of time composed of relatively discrete segments.
behavior during this interaction period took one of our forms:
His
the
communicator was continually positive, continually negative, positive
in the early segments and negative in the remaining segments, or
negative early and positive later on.
The communicator then presented
a standard communication and the subjects' opinions were measured to
determine the effects of these prior statements.
It was found that
the greatest amount of agreement with the communicator was produced
by the communicator who had previously expressed a gain in esteem for
the recipient.
The extent of agreement was next highest in the case
of constant positive esteem, followed by invariate negative esteem,
with loss in esteem producing the least agreement.
The results of a study on bargaining strategies by Deutsch,
Epstein, Canavan, and Gumpert (1967) lend further support to the notion
that negative attitudes expressed toward another followed by the
expression of positive attitudes in inducing cooperation from that
individual.
Their experiment studies five behavioral strategies to
see which was most effective in eliciting cooperative behavior from
someone whose behavior was not initially and persistently cooperative.
The effectiveness.of the strategies was investigated in a two-person
laboratory game which permitted players to act altruistically, coopera­
tively, individualistically, defensively, or aggressively toward one
35
another.
One of the players In each game was always an accomplice of
the experimenter, who followed a predetermined strategy in response to
the true subject's behavior in the game.
The five strategies employed
by the accomplice were termed Turn the Other Cheek, Nonpunitive,
Deterrent, Reformed Sinner-Turn the Other Cheek, and Reformed SinnerNonpunitive.
It was found that the most frequent cooperative responses
were made to the accomplices who employed a Reformed Sinner-Turn the
Other Cheek strategy.
Employing this strategy the accomplice played
in a very threatening and aggressive manner during the first fifteen
trials of the game but then dramatically changed his behavior by
disarming on the sixteenth trial.
He then followed a Turn the Other
Cheek strategy during which he responded to attacks or threats by
altruistic behavior (doing something that rewarded the other) and with
cooperative behavior otherwise.
Summary
The literature revealed that the dimension of warmth in the
behavior of teachers and examiners has a major influence on the
behavior of those with whom they interact.
Thus warmth may be an
important factor in determining a teacher's effectiveness.
Warmth is an important quality in a person which seems to
carry more weight than others in establishing a view of an individual's
personality.
Warmth from others appears to be a genuine need of each
individual and thus constitutes a powerful social reinforcer.
Many
36
studies have demonstrated the effectiveness of minimal verbal and
nonverbal cues of a teacher or an examiner in conditioning a subject's
verbal behavior and in some cases nonverbal behavior.
Some of the
cues which can be construed as indicants of warmth or approval are
verbal responses such as "mmm-hmm," "good," "okay," "yeah," etc.
Nonverbal behaviors shown to be effective reinforcers are headnods
and smiles.
Evidence from several studies supports the hypothesis that
social reinforcement is more effective when it comes from a warm
individual than when it comes from a cold individual.
Furthermore it
appears that a hostile experimenter retards learning in comparison to
a neutral experimenter.
It may be, however, that a teacher or an examiner can behave
too warmly and thus lose his effectiveness as a reinforcer.
It appears
that behaviors maintained by social reinforcers are responsive to a
condition of relative satiation for such reinforcers.
The results of
several studies suggest that a source might be more effective in
producing behavior change in another by expressing different orders of
warmth and/or anger.
The expression of initially negative feelings
toward another followed by the expression of increasingly positive
feelings might be more rewarding to that person than the expression
of invariant warmth.
37
BEHAVIORAL CUES OF WARMTH
The dimension of warmth in the behavior of teachers and
examiners has been cited as having a major influence on the behavior
of those with whom they interact.
It is this wide range of influence
which makes warmth an important variable for investigation.
However,
in spite of the widespread use of the concept of warmth, there has
been no clear definition or specification of the behavioral correlates
of this variable.
Although global measures of warmth utilizing rating
scales have been used in many studies, it is the aim of this section
of the present paper to review those studies which attempt to define
and specify the behavioral cues of warmth.
In a study by Johnson (1971:571), the following definition
was used in training his actors to express warmth.
Warmth can be expressed both verbally and nonverbally.
Verbal expressions such as "That's good," or "That's an inter­
esting thought," are statements of warmth. Nonverbally, warmth
can be expressed through tone of voice, facial expression, and
gestures; for example, leaning toward the other person, looking
directly into his eyes, smiling, and a friendly tone of voice
all communicate warmth. Warmth can mean several different
things. In expressing warmth in this experiment we would like
you to express warmth in a way which means acceptance of the
other person.
Bayes (1972:33) attempted to define interpersonal warmth in
behavioral terms by determining the association between global ratings
of warmth and objective measures of specific behavioral cues obtained
independently.
38
Those cues found to be most closely related to warmth ratings
were (I) frequency of smiling, the single best predictor, and (2)
number of positive statements about other people.
A factor analysis
identified two factors, the first an evaluative one of positive
response to others and the second an activity level factor.
The author
concluded that "warmth may be tentatively defined as positive response
to others, actively conveyed."
After a number of studies performed with college students,
Albert Mehrabian (1968:52) came to this conclusion:
I tell you that feelings are communicated less by the words
a person uses than by certain nonverbal means— that, for example,
the verbal part of a spoken message has considerably less effect
on whether a listener feels liked or disliked than a speaker's
facial expression or tone of voice . . . .
In fact, we've worked
out a formula that shows exactly how much each of these components
contributes to the effect of the message as a whole. It goes
like this: Total Inpact = .07 verbal + .38 vocal + .55 facial.
Through the use of an electronic filter, Mehrabian was able
to measure the degree of liking communicated vocally.
The filter
eliminated the higher frequencies of recorded speech, so that words
were unintelligible but most vocal qualities remain. Mehrabian found
that people were able to judge rather easily and with a significant
amount of agreement the degree of liking conveyed by the filtered
speech.
Given one communication, one group judged the amount of liking
conveyed by a transcription of what was said, the verbal part of the
message.
A second group judged the vocal component, arid a third
group judged the impact of the complete recorded message.
It was
39
found that when the verbal components of a message agree (both posi­
tive or both negative), the message as a whole was judged a little
more positive or a little more negative than either component by
itself.
But when vocal information contradicted verbal, vocal won out.
Some of the other results of Mehrabian's research indicated:
posture is used to indicate liking— the more a person leans toward his
addressee, the more positively he feels about him.
Relaxation of
posture is a good indicator of attitude— a speaker relaxes either
very little or a great deal when he dislikes the person he is talking
to, and to a moderate degree when he likes the person.
Also standing
close to your partner and facing him directly indicates positive
feelings.
In a study of Haase and Teppler (1972:417), twenty-six coun­
selors with an average of 1,500 hours counseling experience rated fortyeight combinations of eye contact, trunk lean, body orientation,
distance, and predetermined verbal empathy message on a modification
of the Truax-Carkhuff empathy scale.
Results showed "that maintaining
eye contact, forward trunk lean, close distance, and medium— and high—
rated verbal empathy all independently contribute to higher levels of
judged empathy
Using photographs of a masked male model as stimuli, James
(1932:405) asked his subjects about the attitude being expressed by
each posture and the portions of the posture which were most
40
significant.
His findings support the hypothesis that a forward lean
communicates a relatively positive attitude (i.e., attentive interest),
whereas a backward lean or turning away communicates a more negative
attitude.
Machotka (1965:33) informally noted relationships between
several postural variables and attitudes.
In his study, drawings of
groups of people who had assumed various postures relative to one
another were judged by subjects who were asked to infer social rela­
tionships.
He found that openness of arms indicates warmth and that
eye contact indicates concern with the addressee.
Argyle and Kendon (1967:74) summarized some of the research
literature to the effects of eye contact in interpersonal communica­
tion.
They report an unpublished study by Weisbrod (1965) who studied
eye contact pattern in a group.
She found that those individuals in
the group who looked most at a speaker were rated by the speaker as
instrumental to his goals and as valuing him more.
She also found
that a speaker feels more powerful when he receives more eye contact
from his addressees.
Furthermore, those individuals who were looked
at most by speakers in the group saw themselves, and were seen by
other group members, as being more powerful in the group than those
who were looked at less.
Another study reported by Argyle and Kendon on the perceptions
of being looked at is that of Mehrabian (Winer and Mehrabian, no date).
41
where the experimenter interviewed two subjects simultaneously, but
spent more time looking at one subject than at the other.
The subjects
were then asked to rate the attitude of the interviewer toward them,
and it was found that the subject who received the most looking judged
the experimenter to be more positive toward her than the subject who
was looked at less.
Three other unpublished studies reported by Argyle and Kendon
lend support to the notion that the degree of eye contact between a
speaker and his addressee is related to the amount of positive regard
conveyed to the addressee.
Kendon (1964) found that subjects thought
that an interviewer who did not look at them for part of the interview
had lost interest in what they were saying.
Exline and Kendon (1965)
found that individuals are judged as more "potent" when they do not
look while the subject is speaking as compared to a condition in which
they do look while he is speaking.
Exline and Eldridge (1965) showed
that subjects judge a speaker as more sincere if he looks at them when
he speaks than if he does not.
The experimental introduction and manipulation of a warmth
variable is incorporated in the design of several studies some of which
are cited elsewhere in the paper.
The effectiveness of the manipula­
tion of the warmth variable in these studies was usually checked by
asking the subjects to rate the degree to which they felt liked or the
degree to which they viewed the person supposedly expressing warmth as
42
a warm person.
In a study by Schmidt and Strong (1971:348), two male graduate
students in counseling psychology, used as interviewers, controlled
the subject’s attraction to them by varying their apparent liking for
and similarity to the interviewee.
In the attractive role the inter-,
viewer, attempting to express a liking for the interviewee, greeted
him warmly, shook his hand, looked and smiled at him, leaned toward
him, responded warmly to him throughout the interview, and indicated
that he liked the same things the subjects liked.
In the unattractive
role, the interviewer ignored the interviewee when he entered the
interviewer's office, did not smile at him, did not look at him beyond
a few odd glances, found no points of similarity with the subject
during the interview, leaned away from him and portrayed disinterest,
coldness, and boredom.
It was found that the subjects who described
the interviewers using a seventy-five item adjective checklist follow­
ing the interview perceived the interviewers as intended.
Mean ratings
for the attractive and unattractive roles differed significantly on
forty-nine of the seventy-five adjective checklist items.
Adjectives
for which mean difference between attractive and unattractive ratings
obtained F ratios exceeding 20 (p .0001) described attractive inter­
viewers as more friendly, good-natured, cheerful, considerate, happy,
warm, attractive and polite, while the unattractive interviewers were
\
described as more cold, humorless, aloof, unhappy, depressed, sad, and
43
selfish.
Strong and Nixon (1971:562) designed a study which was also
successful in varying interviewer attractiveness as evidenced by
subjects' ratings on a seventy-eight item adjective checklist.
In the
attractive role, the interviewer introduced himself by his first and
last name (no title), shook the subject's hand, showed him. into the
office, offered him a chair, and made friendly comments such as "See
you found Student Life Studies all right."
He sat down in his chair,
leaned forward, and moved his chair toward the subject.
He was
responsive to the subject, he looked and smiled at the subject, and
indicated that he liked the same things the subject liked.
At the
end of the interview, he indicated that he had enjoyed talking with the
student by stating, "Ordinarily, these interviews are fairly routine,
but I really have gotten a great deal out of talking with you.
thoroughly enjoyed it.
Thank you!"
I
In the unattractive role, the
interviewer did not greet the subject or discuss any irrelevant topics
before going into the subject matter.
He moved his chair away from
the subject, leaned back in his chair, and did not look at.the subject
except for fleeting, hold glances.
His face was expressionless; he
i.
did not smile; he occasionally covered his face with his hands and
rubbed his eyes to indicate boredom.
He often turned to the side
rather than directly toward the subject.
Subjects' ratings of video-taped interviews provided evidence
44
that counselor attentiveness had been effectively varied in a study
by Krumboltz, Varenhorst, and Thoreson (1967:412).
In the high- y
attractiveness role, the counselor looked directly at the student,
nodding her head or smiling to indicate she was following the conversa­
tion, indicated enthusiasm for the student's thoughts and plans by the
tone of voice.and the expression on her face, and refrained from
distracting mannerisms such as doodling and fidgeting to indicate
pose and competence.
In order to indicate low attentiveness, the
counselor did not smile during the interview, responded to the student
in a flat tone of voice, seldom glanced at the student while she was
talking, frequently rubbed her eyes, played with her hair, or stretched,
and wrote while the student talked and fiddled with objects on the
desk.
It was found that subjects who had observed the high-attentive
counselor rated her significantly more interested in the student's
conversation than did those who observed the low-attentive counselor.
Warmth of interaction was one of the variables systematically
studied in an experiment by Johnson (1971:20.7) which investigated the
process of negotiations and the induction of cooperation in a negotia­
ting situation.
Warmth of interaction was experimentally manipulated
by training confederates to negotiate and role reverse in ways which
expressed either warmness or coldness.
They were trained to express
warmth or coldness in their tone of voice and facial expression, to
lean toward the subjects when expressing warmness, and to look into the
45
subject's eyes when expressing warmth and to avoid looking into the
subject's eyes when expressing coldness.
It was found that the expres­
sion of warmth, by means of these behavioral cues, compared with the
expression of coldness, resulted in more favorable attitudes toward
the confederates as a more understanding person, as more accepting of
their position and of them as persons, and as being more similar, both
as a person and in beliefs and values.
Subjects in the warm conditions
also liked and trusted the confederate more.
Summary
The review of the literature relating to the specific
behavioral correlates of warmth indicated that warmth can be expressed
both verbally and nonverbally, and that the verbal part of a spoken
message may have considerably less effect on whether a listener feels
liked or disliked than certain nonverbal cues exhibited by a speaker.
The results of several studies indicated that looking at
someone while they are speaking, smiling, and leaning forward when
seated, portray warmth and attentiveness to that person.
An individu­
al's tone of voice and facial expression while addressing someone can
indicate liking for that person.
There is also some evidence that
maintaining a moderately relaxed posture, standing closer to your
partner, and facing him, while speaking or listening to him, communi­
cate positive feelings.
Also, making positive statements about other people may
46
contribute to one's being perceived by others as a warm person.
PRESTIGE AND ITS EFFECT UPON LEARNING
The studies reviewed so far have indicated that warmth or
approval is a string social reinforcer and is thus an important
aspect of the teacher-learner interpersonal relationship.
These
studies have used minimal verbal and nonverbal cues of the examiner
such as "mmm-hmm" or headnodding on the basis that such cues indicate
attention and interest, are quite natural and realistic, and are more
effective than generally realized (Krasner and Ullman, 1965).
Krasner and Ullman (1965:15) stated:
At first, it was thought by investigators that "mmm-hmm"
or "good" could be set up as objective types of responses,
analogous perhaps to food pellets, which could be delivered
in certain quantities with specifiable schedules. However,
studies thus far have clearly indicated that the reinforce­
ment could not be divorced from the "giver" of the reinforce­
ment. Unlike animal studies, the magazine delivering the
pellet is a crucial variable.
Krasner (1962:88) further explained:
. . . if the reinforcer is defined as a contingent stimulus,
the person dispensing the reinforcement is a significant aspect
of that stimulus configuration. That is, the experimenter or
teacher, when he uses his own behavior as the reinforcing stimu­
lus, is an important and inseparable part of the reinforcement
machine.
Riley, Ryan, and Lifschitz (1950:193) asked college students to
state ideal factors important in teaching as well as factors which
played a part in the actual teaching they experienced.
\
Personality
47
was frequently mentioned, both as an ideal factor for effective teach­
ing and as a characteristic of their best teachers.
Maslow and
Zimmerman (1956:185) asked students to make ratings of their teachers'
ability and personality on a scale ranging from very good to very poor.
They found that the correlation between student ratings of good
teaching and good personality was r=.76.
Tedeschi, Bonoma, and Schlenker (1972:26)
have proposed a
general theory of social influence within dyads which states that the
reinforcing strength of a source's behavior in influencing the behavior
of another depends on the way he carried out his promises to reinforce,
and his threats to punish different behaviors of the target in the
past.
Given this information, a probability statement can be made
regarding the influence the source's threats, promises, warnings, or
mendations will have on the target's behavior in the future.
Tedeschi, Bonoma, and Schlenker report on a study by Helm,
Brown, and Tedeschi (1972) which briefly outlined the Subjective
Expected Value (SEV) theory of social influence.
. . . a threat specifies a source demand and indicates
the source's intention to punish the target for noncompliance.
Similarly, a promise presents a source's request and offers a
reward for noncompliance. The proportion of times the source
has actually punished noncompliance to this threat or has
rewarded compliance to his promises in previous interactions with
the target defines the probability component of current threats
or promises. The actual magnitude of punishment or reward stipu­
lated in the current message defines the value associated with
the influence attempt. The relationship between these two
components is assumed to be multiplicative, yielding the
expected value (EV) of a threat or a promise. All else equal,
target compliance to promises is assumed to be a direct function
of expected value, whereas compliance to threats is directly
48
mediated by the expected costs of noncompliance.
The theory briefly outlined above also postulates that source
characteristics of status, esteem, prestige, and attraction cause
the target to bias estimates the probabilities associated with
the various message types. These biasing factors lead to sub­
jective expected value (SEV) considerations, since they cause the
target individual to behave in a manner which cannot be predicted
by expected value considerations alone. In a sense "irrational"
conduct is specified and predicted by Tedeschi1s SEV theory of
social influence.
The SEV theory of social influence postulates that the pres­
tige of the source and positive attraction for the source should cause
the target to exaggerate the probability estimations made of low
credibility promises,. whereas low prestige of the source and negative
attraction should cause the target to underestimate the probability of
contingent rewards associated with highly credible sources.
Hovland, Janis, and Kelly (1953:20) point out:
If a communicator is personally admired or a member of a
high status group, his words may raise the incentive value of
the advocated opinion by suggesting that approval, from himself
or from the group, will follow its adoption. These feelings of
affection and admiration from the audience may stem in part from
desires to be like him.
In some studies which investigated the effect the status of
the source has on his influence over others, experimental manipulation
of the status variable involved varying the dress and general appear­
ance of the source.
The initial assumption of these studies is that
one’s physical appearance and manner of dress are important determin­
ants the extent to which one is perceived by others as prestigious.
Describing social techniques used by an individual to present an image
49
of prestige and competence, Argyle and Kendon (1967:82) state:
To create perceptions of and attitudes toward the self on
the part of others present is a subtle social skill, though one
that is usually practiced quite unconsciously . . . .
Another
method of projecting an identity.is by means of clothes and
general appearance, which are in fact excellent clues to a
person’s self-image.
The results of a study by Lefkowitz, Blake, and Mouton (1955:
704) showed that pedestrians violated the prohibition of an automatic
traffic signal more often in the presence of an experimenter's model
who violated the prohibition than
absent.
when the latter conformed or was
Significantly more violations occurred among pedestrians
when the nonconforming model was dressed to represent high social
status than when his attire suggested lower status.
The clothing worn
by the experimenter's model, a thirty-one-year-old male, intended to
typify a high status person was a freshly pressed suit, shined shoes,
white shirt, tie, and a straw hat.
Well-worn scuffed shoes, soiled
patched trousers, and an unpressed blue denim shirt served to define
the model as a low status person.
A study by Menard (1972:3394) carried out in an actual
teaching situation did not produce evidence indicating a relationship
exists between teacher effectiveness and teacher appearance.
In
addition to investigating this relationship, Menard attempted to
determine if the student characteristics of sex, major, achievement,
and socioeconomic status are predictors of teacher effectiveness based
upon a difference in teacher appearance.
The only difference in the
50
way students were taught during two different quarters was in the
appearance of the instructor.
During winter quarter, the instructor
had long hair, a full beard, and was dressed in faded blue jeans, a
work shirt, and boots.
During spring quarter, the same instructor,
teaching the same course, had short hair, was clean shaven, and was
dressed in a white shirt and tie, dress slacks, and dress shoes.
Menard concluded that:
There was no difference in teacher effectiveness as measured
by student ratings or student gain regardless of the appearance
of the teacher and that the student characteristics of sex,
major, achievement, and socioeconomic status did not aid. in the
prediction of teacher effectiveness (1972:3395).
Mills and Aronson (1965:173) found that a physically attractive
communicator was more effective than an unattractive one, but only if
she expressed a desire to change opinions.
In the attractive condi­
tions, a female communicator was made up to look physically attractive.
"She wore chic, tight-fitting clothing; her hair was modishly coiffured;
she wore becoming makeup."
In the unattractive condition the same
communicator was made up to look repulsive.
"She wore loose, ugly,
ill-fitting clothing; her hair was messy; her makeup was conspicuously
absent; the trace of a moustache was etched on her upper lip; her
complexion was oily and unwholesome looking."
Subjects' ratings of the
communicator's characteristics provided strong evidence that she was
perceived as more attractive in the attractive conditions than in the
unattractive conditions.
It is of interest to note that in the
51
attractive condition she was also rated, to a significant degree, more
charming and more affectionate.
The data indicated that the attractive
communicator was more effective in producing opinion change but only
if she expressed a desire to change opinions.
The purpose of a study by Haiman (1949:192) was to determine
whether a combination of two specific factors only— a speaker's
likeableness and physical attractiveness— would influence the effective­
ness of his persuasion, as shown by a difference or lack of difference
in audience shifts of opinion pursuant to a variation in those
factors.
In the first mode, the speaker tried his best to make a
favorable impression upon the audience as regards the factors of
physical appearance and likableness.
In the second mode, the
speaker attempted to, be just as effective in delivery and general
competence, but to make an unfavorable impression upon the audience
as regards physical attractiveness and likableness. This latter
was to be accomplished by failing to shave for about 24 hours
before the speech, failing to comb his hair, wearing an extremely
unattractive pair of glasses, wearing a just noticeably dirty
white shirt, slightly torn at the sleeve, and a very un-neatIy
tied tie, wearing a dirty and completely unpressed pair of
trousers, wearing scuffed and unshined shoes, an air of super­
ciliousness, sarcasm, and unfriendliness toward the audience
(1949:194).
Results indicated that an opinion shift occurred in the
predicted direction which was not quite statistically significant.
Several studies have demonstrated the operation of a socalled prestige factor on the effectiveness of a communication.
That
is, a prestigious communicator is better, able to modify the behavior
of others than is a source of low prestige.
Tedeschi, Bonoma, and
52
Schlenker report an experiment by Helm, Brown, and Tedeschi (1972) which
was designed to test the effects of experimenter expertise upon the
performance of subjects in a verbal reinforcement task.
The experimen­
ter was represented as a doctoral candidate collecting his dissertation
data or else as an undergraduate fulfilling an assignment for his
sophomore level experimental psychology course.
Following their task,
the subjects were asked to complete an Interpersonal Judgment Scale
(Byrne, 1969:35), which asked for the subject’s evaluation of the
experimenter and from which scores were obtained concerning inter­
personal attraction and esteem.
It was found that subjects rated the
more expert experimenter as more respected and intelligent (i.e.,
I
esteemed) than did subjects who rated the less expert experimenter,
thereby supporting the effectiveness of the esteem manipulation.
The
reward offered by the experimenter was social approval— he said "good"
!
when the desired response was emitted— and when it was offered by the
esteemed experimenter it produced more reinforced responses (i.e.,
compliance to the source's tacit requests).
A work by Oakes (1962:469) illustrated the importance of
prestige, in that it compared the effectiveness of signal.light rein­
forcers given various .meanings on the verbal behavior of members
group discussion.
pf
Although the results were not statistically
significant, it is noteworthy that the trends were in the expected
direction.
The most effective reinforcer tested in this situation
seemed to be the light signifying that the subject's statement
a
53
indicated insight or lack of it as judged by statements made by the
professional team that originally worked with the patient, while the
least effective was the light signifying the subject's statement
agreed or disagreed with statements made by a group of laymen who had
previously discussed the case.
Mausner (1953:391) tested the hypothesis that subjects will
be influenced more by the judgments of a partner with high prestige
than by those of one with low prestige.
In this study, three groups
of ten subjects, equated for interest in art by means of the AllportVernon Scale of Values, were given the Meier Art Judgment Test.
Subjects in Group I repeated the test alone; subjects in Group II and
Group III repeated it with a partner.
He was introduced to Group II
as a fellow student; to Group III subjects as an "art authority."
In
the together situation, both members of the pair judged each of the
pairs of the pictures; the subject was first in all even trials; the
confederate in all odd trials.
The confederate had memorized the
test; he consistently stated the preference indicated as wrong by the
scoring key.
Degree of social influence was measured in terms of the
shift in frequency of wrong judgments from the alone to the social
i
situation.
It was found that Group I (control) showed no significant
shift in judgments, while both Groups II and III did show a significant
shift with Group I^I (art authority) showing a significantly greater
shift than Group II (fellow student).
54
Kulp (1934:663) administered a test containing seventy-one
propositions dealing with social, economic, political, religious,
international, and educational problems to a large group of graduate
students in educational sociology.
At the second administration of
the same test, which occurred one week after the first test, one
group of subjects were told that a carefully selected group of social
scientists, as experts, had passed judgment on each proposition, a
second group was told graduate educators from ten outstanding schools
had passed judgment, while a third group was told that a large number
of lay citizens had passed judgment.
A fourth group received either
liberally or conservatively marked test blanks with no statement.
A
control group received unmarked test blanks as they did during the
first testing.
In this way a prestige variable was introduced in
order to produce the desired changes in attitudes, either toward a
more liberal or conservative point of view as defined by the author
of the test.
Results indicated that a significant shift in attitudes
was effected by manipulating suggestion and prestige, with the greatest
amount of prestige enjoyed by educators as authorities, and the next
greatest amount enjoyed by graduate students in education.
A study by Das (.1960:487) indicated that the depth of response
to suggestion is greater when the source is a prestigious person.
Tape-recorded body-sway suggestions by the Head of the Pyschology
Department were significantly more effective than those of less
i'
55
prestigious sources in influencing body-sway in subjects who were
high in suggestibility.
Several studies have produced evidence indicating that high
status people are more successful than low status people in shaping
the opinions of others.
In a study by Goldberg and Iverson (1965:
673), subjects first completed a questionnaire dealing with various
aspects of health and nutrition and were then assigned partners of
either high or low status (confederates of the investigator) with whom
they listened to three taped speeches expressing conventional or
unconventional viewpoints on the same topic.
At the end of each
speech, the partners first filled out an opinion questionnaire and
then showed their responses to the subjects.
The subjects followed
by responding to these items for a second time.
In order to introduce
the status variable the confederate was presented to half of the
subjects as a second-year medical student who had formerly graduated
from the university with honors (high status) and to the other half as
a hospital orderly who had not completed high school (low status).
The amount of the confederates' influence on the subjects initial
ratings and those recorded later.
It was found that the high status
confederates had a significantly greater amount of influence than the
low status confederates.
In an investigation by Haiman (1949:192), the prestige of the
speaker was experimentally varied by conveying to the audience
56
knowledge of his character and reputation prior to.his speech and also
by a chairman's introduction.
After marking their opinions on a
questionnaire, three different audiences heard the same tape-recorded
speech.
The members of one audience were told that the recording they
were about to hear was of a speech by Eugene Dennis, Secretary-General
of the Communist Party of America.
Members of the second audience :
were told that the speaker was Dr. Thomas Parran, Surgeon General of
the United States, and were told, in a short paragraph, about Dr.
Parran's impressive professional background.
The members of the third
audience thought it was a speech by an anonymous college sophomore.
Following the speech, the audiences marked after-speech opinions, and
also filled out a rating scale which confirmed that the persons
selected by the experimenter to represent different degrees of prestige
actually did differ in the minds of the audience.
It was found that
the shift of audience opinion obtained by Dr. Parron, who was highest
in prestige as indicated by student ratings, was significantly greater
than that obtained by either Mr. Dennis or the Northwestern sophomore.
The difference between Mr. Dennis and the sophomore was not significant.
Lorge (1936:402) designed a study that tested the hypothesis
that a person, confronted with an opinion from one who has prestige
for him, will have his reaction to it colored accordingly.
He
attempted to measure quantitatively the capability of the factor of
prestige to alter the evaluations of statements concerning serious
57
political and economic questions.
On two occasions, two weeks to a
month apart, subjects rated each of a set of fifty brief quotations
on a five-point scale, indicating the degree of his agreement (or
disagreement) with them.
It was found that subjects tended to rate
the same statement differently when it was attributed to a different
author.
More specifically, the changes in the ratings of the state­
ments corresponded in direction to the differences in the subject's
ratings of the authors.
Kelman and Hovland (1953:327) also found evidence that the
prestige of the communicator was important in regard to his ability
to produce opinion change.
The prestige factor was experimentally
manipulated by introducing one speaker as "Judge Howard Elson, presid­
ing judge of the Juvenile Court of this city, author of several books
on delinquency, and well known for his views on the integration of the
delinquent into society," while introducing the other speaker as a
"man on the street" picked from the studio audience.
Furthermore, the
prestigious communicator attempted to present a more trustworthy, wellinformed image while the low-prestige communicator presented an
untrustworthy, poorly informed image.
The intended difference in the
perception of the communicators were achieved, as indicated by pro­
nounced and statistically significant differences in the students'
appraisal of the competence, fairness, and trustworthiness of the
communicators.
Results indicated a greater effect of the communication
58
on the opinions of the students was achieved by the more prestigious
and trustworthy communicator.
The "findings of a study by Aronson, Turner, and Carlsmith
(1963:31) lend further support to the notion that a positive relation­
ship exists between the prestige of the communicator and the extent of
opinion change.
This experiment investigated the interaction between
the credibility of the communicator and the discrepancy of the communi­
cations supposedly written by a highly credible source— an expert on
poetry.
A control group read virtually identical essays, supposedly
written by a student.
It was found that subjects who read a communica­
tion that was attributed to a highly credible source showed greater
opinion change when the opinion of the course was presented as being
increasingly discrepant from their own.
In sharp contrast to this was
the behavior of subjects who were exposed to the same communication—
attributed to a source having only moderate ability.
In this condition
increasing the discrepancy increased the degree of opinion change only
to a point; as discrepancy became more extreme, however, the degree of
opinion change decreased.
A study by Bochner and Insko (1966:614) also found opinion
change to be linearly related to communicator— communicatee discrepancy
for a high credibility source, and curvilinearIy related to communica­
tor-communicatee discrepancy for a medium credibility source.
Communicator credibility was manipulated by attributing the same
I
59
communication to either "Sir John Eccles, Nobel prize winning physiolo­
gist," or to "Mr. Harry J. Olsen, director of the Fort Worth Y.M.C.A."
Browning (1965:4803) found that clients counseled by high
prestige therapists accepted discrepant interpretations by the thera­
pist and maintained a more positive relationship with him when
presented with such interpretations than did clients of low prestige
therapists.
Among Browning's conclusions were:
Subjects counseled by a low prestige therapist tend to reject
highly discrepant interpretation more so than those associated
with a high prestige therapist.
Subjects under low prestige conditions receiving discrepant
interpretations in the sequence, moderate, high and low, indicate
significantly less counselor-subject rapport than subjects in
the high prestige condition receiving the same sequence (1965:
4804).
Bergin (1962:423) experimentally manipulated therapist credi­
bility by having one group of subjects report individually to the
Psychiatry Department of the Stanford Medical Center where the experi­
menter assumed the role of director of a personality assessment project.'
To further establish his credibility, subjects were sent to the
experimenter by a receptionist, and the experimental room was furnished
with elaborate equipment, a couch, an impressive array of medical and
psychological volumes, and a large portrait of Freud.
In the low
credibility conditions, subjects reported for the first session to a
decrepit room in the basement of the Education Building.
Bergin found
that subjects receiving a communication from a source of high
'I
60
credibility changed their self-ratings in the direction of the commu­
nication from a low credibility source.
Sarason and Minard (1963:87) used a verbal conditioning
paradigm with first-person pronouns as the reinforced response class
in testing the effects of experimenter hostility and prestige on the
subject's performance.
Subjects in the high-prestige condition were
greeted in the experimental situation by a business-like, well-dressed
experimenter whose name was on the door of the room.
Subjects in the
low-prestige condition were greeted by a casually dressed student who
said, "I guess you're mine."
High and low hostile experimenters, as
well as high and low hostile subjects, were determined by their scores
on the Hostility scale of Sarasons' (1958:339)
Autobiographical Survey
which had been administered prior to, and independent of, the study.
It was found that of the subjects run by high hostile experimenters
only those who were run under the high-prestige condition showed an
increase in responses of the reinforced class.
Subjects run under the
low hostile experimenter-low prestige condition also exhibited a
learning effect.
Strong and Nixon (1971b :562) tested the hypothesis that
expertness masks the influence of attractiveness.
expertness, attractiveness affects influence power.
That is, with '
In order to
experimentally manipulate the attractiveness variable, the counselor
interviewed one-half the subjects in a warm, friendly manner, and
61
one-half in a cold, unfriendly manner.
Furthermore, the interviewers
were introduced to one-half the subjects as inexpert, and to oneyhalf
as expert.
The inexpert introduction was "I'm sorry . . . but Dr.
______________ called and said he couldn't make it . . . Dick Williams,
a first-year student here, said he would try it.
He has little back­
ground or experience, but he does have a vague idea of the purpose of
the study (1971:
)."
The nameplate on the desk was removed.
The
expert introduction included the Dr. reference and the Dr. ___________
nameplate.
Thus, four experimental conditions were created, expert-
attractive, expert-unattractive, inexpert-attractive, and inexpertunattractive.
Interviewer influence was measured by the amount of the
subjects' shift in opinion at three levels of discrepancy between the
subjects' initial self-estimates of their overall achievement motiva­
tion and the interviewer's opinion.
The results indicated that expert
interviewer's attractiveness does not affect their influence power,
while inexpert interviewer's attractiveness defines their influence
power.
It was also found that a masking effect of expertness occurs
in the unattractive role conditions.
Expertness carries the inter­
viewer's influence in spite of the negative effects of his unattractive­
ness.
Commenting on the greater degree of hostility that was expressed
toward the inexpert-unattractive interviewer in comparison to the
hostility expressed toward the expert-unattractive interviewer, the
authors remarked, "Obviously, if one must be an unattractive,
i
62
discourteous fellow, one had better be an expert.''
Learning can occur vicariously through observations of the
behavior of social models.
Thus, modeling can be an effective proce­
dure for transmitting and controlling behaviors.
Social models who
are perceived as attractive, prestigeful, competent, and high in
status result in increased imitative behaviors by observers (Thoreson
and Krumboltz, 1968:393).
Bandura and Walters (1963:195) contend:
No doubt every teacher employs modeling as one of his
techniques, whether consciously or unconsciously. His potency
as a model will be influenced by such characteristics as age,
sex, socio-economic status, social power, ethnic background,
and intellectual and vocational status.
In regard to a counselor's effectiveness in influencing the
,
behavior of his client, Krumboltz, Varenhorst, and Thoreson (1967:412)
make the following statement:
The success of a counselor in using reinforcement and modeling
procedures may depend upon the amount of regard or prestige
attributed to him by the client. If the counselor is perceived
as prestigeful or if the social model is seen as socially power­
ful, the counselor may be more influential in modifying relevant
behaviors. A variety of possible cues may be used by the client
in arriving at an estimate of the counselor's prestige or social
power. Information about the experience, training, respect, and
effectiveness as reported by other people would seem relevant to
the client's perception.
Thoreson and Krutoboltz (1968:393) designed a study to produce
information bearing on the question "Does the success level of peer
'
;
social models affect their influence on students of differeing
abilities?"
Audio tape-presented models depicting three levels of
63
athletic success discussed vocational planning activites for fortyeight eleventh-grade males self-rated as to high, medium, or low
athletic success. A similar three-by-three design involved academic
success with seventy-two males.
The results showed that:
(a) differ­
ent athletic-model success levels caused significant differences in
frequency of information-seeking behaviors by students; (b) the highsuccess athletic model was most effective for all students; and
(c) variations in academic-model success levels did not produce
significant differences in information seeking.
Walls and Smith (1970:123) investigated the possible relation­
ship of model status to the voluntary delay of reinforcement behavior
of adult vocational rehabilitation clients.
Adult vocational rehabili­
tation clients in a resident training program were engaged in a task
involving the assembly of bolts, washers, and nuts and were given their
choice of a smaller reward immediately or a delayed larger reward.
As
an obstensibly incidental portion of the procedure, subjects were
permitted to view one of four video tapes (immediate or delay decisions
by a medium-status, peer or a high-status counselor).
It was found that
subjects, regardless of disability and aptitude, tended to imitate
the decision of the video-tape model (both high and medium status) to
which they were exposed.
The effect of information concerning the competence of a model
on learning of imitative and nonimitative behavior was examined by
64
Rosenbaum and Tucker (1962:183).
The subjects predicted the outcomes
of a series of fictitious horse races after exposure on each trial to
the prediction and correctness of the prediction made by a simulated
partner.
The competence of the model, which referred to the degree
of correctness of his predictions, was the independent variable.
Under
training to imitate— a response by the subject matching that of the
model was correct— the results indicated that the greater the model's
competence, the greater was the facilitation of the learning process.
Under training to nonimitate— a response by the subject matching that
of the model was incorrect— no differences appeared among three condi­
tions varying the model's competence.
Summary
The literature relating the prestige of a source to his
ability to modify the behavior and attitudes of others indicated that
the effectiveness of a communication in producing opinion change
depends to a large degree on the extent to which the communicator is
held in esteem by the recipient of the communication.
Similarly, the
effectiveness of social reinforcement depends a great deal upon the
prestige of the dispenser of that reinforcement.
Information about a source's experience, training, and
competence as reported by other people seems relevant to the amount of
prestige that is attributed to him.
Physical appearance and manner of
dress are also important determinants of the extent to which one is
65
perceived as prestigious.
There is much evidence that a person confronted with an opinion
from one who has prestige for him will have tiis reaction to it colored
accordingly.
Many studies in the area of social influence have shown
that people of high prestige are more successful than people of low
prestige in shaping the opinions of others.
The results of several
other studies indicated that the effectiveness of social approval in
modifying behavior is positively related to the prestige of the person
delivering the approval.
Also, an individual's potency, and intellec­
tual, economic, and vocational status.
There is some evidence that if
a source is seen as an expert, the degree of warmth he exhibits has
little effect upon his influence power.
SUMMARY
This chapter contained a review of selected literature relating
to the personal characteristics of warmth and prestige and the role
i
they play in interpersonal communication, especially as they pertain
to the teaching-learning process.
Another focus of this chapter was
the specific behaviors that convey warmth, that is, those behaviors
which tell another that .he is regarded in a positive way.
This chapter was organized in the following sequence:
(I) the
influence of warmth <?n the effectiveness of communication, (2) behavi­
oral cues of warmth, and (3) the influence of source prestige on the!
66
effectiveness of communication.
The summary of the literature revealed that the dimension of
warmth in the behavior of teachers and examiners has a major influence
on the behavior of those with whom they interact.
Thus, warmth may
be an important factor in determining a teacher's effectiveness.
Warmth is an important quality in a person which seems to
carry more weight than others in establishing a view of an individual's
personality.
Warmth from others appears to be a genuine need of each
individual and thus constitutes a powerful social reinforcer.
Many
studies have demonstrated the effectiveness of minimal verbal and non­
verbal cues of a teacher or an examiner in conditioning a subject's
verbal behavior and in some cases nonverbal behavior.
Some of the
cues which can be construed as indicants of warmth or approval are
verbal responses such as "mmm-hmm," "good," "okay," "yeah," etc.
Non­
verbal behaviors shown to be effective reinforcers are headnods and
smiles.
Evidence from several studies supports the hypothesis that
social reinforcement is more effective when it comes from a warm
individual than when it comes from a cold individual.
Furthermore,
it appears that a hostile experimenter regards learning in comparison
to a neutral experimenter.
It may be that a teacher or an examiner can behave too warmly
and thus lose his effectiveness as a reinforcer.
It appears that
67
behaviors maintained by social reinforcers are responsive to a
condition of relative satiation for such reinforcers.
The results
of several studies suggest that a source might be more effective in
producing behavior change in another by expressing different orders
in warmth and/or anger.
The expression of initially negative feelings
toward another followed by the expression of increasingly positive
feelings might be more rewarding to that person than the expression
!
of invariant warmth.
The literature relating to the specific behavioral correlates
of warmth indicated that warmth can be expressed both verbally and
nonverbally, and that the verbal part of a spoken message may have
considerably less effect on whether a listener feels liked or disliked
than certain nonverbal cues exhibited by a speaker.
The results of several studies indicated that looking at
somone while they are speaking, smiling, and leaning forward when
seated, portray warmth and attentiveness to that person.
An individu­
al's tone of voice and facial expression while addressing someone can
indicate liking for that person.
There is also some evidence that
maintaining a moderately relaxed posture, standing closer to your
partner, and facing him while speaking or listening to him communicate
positive feelings.
Also, making positive statements about other people may
contribute to one's being perceived by others as a warm person.
I
68
The literature relating the prestige of a source to his
ability to modify the behavior and attitudes of others indicated that
the effectiveness of a communication in producing opinion change
depends to a large degree on the extent to which the communicator is
held in esteem by the recipient of the communication.
Similarly, the
effectiveness of social reinforcement depends a great deal upon the
prestige of the dispenser of that reinforcement.
i
Information about a source's experience, training, and compe­
tence as reported by other people seems relevant to the amount of
prestige that is attributed to him.
dress are also important to him.
Physical appearance and manner of
Physical appearance and manner of
dress are also important determinants of the extent to which one is
perceived as prestigious.
There is much evidence that a person confronted with an opinion
from one who has prestige for him will have his reaction to it colored
accordingly.
Many studies in the area of social influence have shown
that people of high prestige are more successful than people of low
prestige in shaping the opinions of others.
The results of several
other studies indicated that the effectiveness of social approval in
modifying behavior is positively related to the prestige of the person
delivering the approval.
Also, an individual’s potency as a social
model is influenced by such characteristics as competency, and intellec
tual, economic, and vocational status.
There is some evidence that if
69
a source is seen as an expert, the degree of warmth he exhibits has
little effect upon his influence power.
Chapter 3 describes the procedures carried out in this inves­
tigation.
It includes descriptions of the community and population,
methodology and types of data collected, hypotheses, and how the data
was analyzed.
i ■
Chapter 3
PROCEDURES
It was the intent of this study to determine the relationships
among selected teacher behaviors and characteristics and certain
measures of teacher performance.
This chapter was developed around the following outline:
1.
A description of the community
2.
A description of the population <
3.
The sampling procedure
4.
Types of data collected
5.
Method of collecting data
6.
Hypotheses
7.
Analysis of the data
'7
J
"
COMMUNITY DESCRIPTION
Montana State University was the first state institution of
higher education to open in Montana.
It was established by the legis­
lature, as the state's agricultural land-grant institution in 1893.
Having begun as a new school with land but no buildings, the
Montana State University campus now extends over 1,170 acres and has
more than forty major buildings.
It is located on the outskirts of
!
the city of Bozeman (population 18,000), which is situated in the
Gallatin Valley, a rich ranching and farming area.
■
71
Montana State University has grown into a multipurpose insti­
tution consisting of five colleges:
the College of Agriculture,
College of Education, College of Engineering, College of Letters and
Sciences, and the College of Professional Schools (including the
schools of Art, Commerce, Home Economics, Nursing, and the departments
of Film and Television Production and Music).
Bachelor's degrees are
offered in more than forty-five fields with some one-hundred-nineteen
separate majors, master's degrees in thirty difference areas, and
doctorates in nineteen.
Montana State University's environment is a great attraction
to faculty members, some of whom are highly regarded in their
fields and could have taught at more prestigious institutions.
Fifty-five percent of the faculty have doctorates. Most are
relatively young, in the 35 to 45 age bracket. M.S.U. doubled
in size in the mid-sixties and its greatest strength lies in
the associate professor category (1975:13).
DESCRIPTION OF THE POPULATION
The population of this study consisted of all the 211 junior
level courses during Spring Quarter, 1975, as part of the curricula of
the five Colleges at Montana State University (see Table 2, page 72,
for the population distribution by College).
Of the forty-seven instructors asked to participate in this
study, forty-two responded affirmatively while fiye responded nega­
tively.
The last six instructors who responded affirmatively were not
used in the study because the number of students who volunteered to
72
be raters was thirty-six.
Table 2
Population Distribution by College
College
300 Level Courses
Offered Spring
Quarter 1975
Courses Selected
for This Study
Agriculture
19
6
Education
37
7
Engineering
75
17
Letters and Science
24
2
Professional Schools
42
4
The raters used in this study were volunteers from three
sections of Educational Psychology 208.
These students were offered
extra credit by the instructors of these classes for their participa­
tion.
The seventeen raters used in validating the Teacher Behavior and
Characteristic Checklist (TBCC), an instrument developed by the
investigator and used in this study, were also volunteers from an
Educational Psychology 208 class.
Validation of this instrument took
place during Winter Quarter, 1975.
SAMPLING PROCEDURE
A list of all the 300 level courses offered during Spring
i
Quarter 1975 was complied and a number was assigned to each course.
73
Numbers were then selected from a random numbers sheet until eighty
courses had been designated.
Laboratory courses were excluded from
the study, because it was reasoned that teacher-student verbal and
nonverbal interactions would not be as readily observed by all the
members of a class in a laboratory situation as is the case in a
regular classroom setting.
The instructors of the courses selected,
according to the, order in which their courses were chosen, were then
asked to participate in the study.
If an instructor did not want to
participate, and five instructors did not, the next person on the list
was contacted.
This procedure was followed until thirty-six instruc­
tors were obtained.
In this way, the courses included in the study
were selected in such a manner that every course in the population had
an equal chance to be chosen (Guilford and Fruchter, 1973:122).
TYPES OF DATA COLLECTED
Data pertaining to teacher behaviors and characteristics
exhibited in the classroom and measures of teacher warmth, teacher
prestige, and teacher effectiveness was gathered on all of the sub­
jects participating in the study.
Teacher behaviors and characteristics were measured by the
Teacher Behavior and Characteristics Checklist, which was developed by
the investigator.
Data relating to teacher warmth, prestige, and
effectiveness was obtained from teacher self-ratings and the
74
administration of the Student Perceptions Scale.
This scale is
comprised of three instruments, the Authoritativeness■Scale, a
modified version of the Scale for Measurement of Counselor Traits,
and the Revised Faculty Rating Form.
Teacher Behaviors and
Characteristics
Instructor behaviors and characteristics were counted and
categorized according to the Teacher Behavior and Characteristic
Checklist, an instrument developed by the investigator.
Following
are the eleven categories included in this scale, divided into two
classes, along with a description of each category.
Class I .
1.
j
Measures of Instructor Prestige
Status and credits:
and credits of the instructor.
This category refers to the status
It includes such information as the
instructor's title, position, books he has had published, and awards
he has received.
2.
Dress:
This category includes a notation of the instruc­
tor's manner of dressing.. It has as its subcategories suit, sports
coat with dress pants, shirt and tie with dress pants, dress pants with
shirt or sweater and no tie, and other, for male instructors.
Data
relating to this dress of the five female teachers who participated
in the study was not collected.
3.
Relation of positive personal experiences:
This category
75
is concerned with the number of times an instructor discusses the
places he has visited (outside of Montana), positions or jobs he has
held, or the degrees or awards he has received.
Also counted are any
references to places he is definitely going to visit in the future,
definite position and job offers he has received, and any degree
programs he is presently in.
References to the positive personal
experiences of his immediate family are also included in this category.
Class II.
I.
Measures of Instructor Warmth
Student utterances:
This category includes all student
responses addressed to the instructor that are not questions unless a
question is followed by an instructor statement such as "That’s a good
(dumb) question."
A question would also be counted as an utterance if
the instructor were to sigh, purse his lips, or raise his voice in
disgust in response to the question;
Although normally considered a question, a statement made with
the student raising his voice at the end is counted as an utterance.
For example, "Trees are green?" is counted as an utterance, while "Are
trees green?" is not.
Also, a student response that begins with a
question but ends with a statement is considered an utterance.
example, "Are trees green?
For
I've seen only red ones," is an utterance.
Not counted as an utterance is a student statement which is
interrupted by, or immediately followed by, another student's statement
or question.
However, if the latter is a statement and is addressed to
76
the instructor, it is considered an utterance.
A verbal exchange consisting of more than one response on the
part of the student and/or the instructor is counted as one utterance.
Such an exchange is considered complete when another student makes a
statement or asks a question, when an instructor or a student state­
ment is fifteen or more seconds in duration, or when a silent period
of fifteen or more seconds follows an instructor or student statement.
2.
Positive evaluation of student utterance;
This category
includes the number of times the instructor responds to student
utterances in a positive way.
Instructor behaviors included in this
i
category are headnodding, smiling, referring to the student by name,
or saying "good," "yeah," "mmm-hmm," "okay," "That's a good point," "I
agree," etc., either during or after a student utterance.
Also
included are instructor paraphrases of student utterances.
This category can be marked a maximum of one time for each
student utterance.
Also, a response such as "That’s a good question"
is included in this category.
If the instructor exhibits both category
2 and category 3 behaviors during or after a student utterance, only,
the last behavior exhibited is counted.
3.
Negative evaluation of student utterances: This category
includes the number of times the instructor responds to student
utterances in a negative way.
Instructor behaviors included in this
category are headshaking, saying "no," "wrong," "uh-uh," "That’s
77
wrong," sighing without smiling, and closing eyes or pursing lips
without smiling.
This category can be marked a maximum of one time for each
student utterance.
Also, a response such as "That's a dumb question,"
is included in this category.
If the instructor exhibits both
category 2 and category 3 behaviors during or after a student utter­
ance, only the last behavior exhibited is counted.
4.
Acknowledgement of student feelings and opinions:
If the
instructor inquires into the students' opinions and feelings, these
behaviors are coded 4.
Not included are purely academic questions
having only one correct answer, such as "What answer did you get for
the third problem?"
This category also included the listening skills of reflecting
feelings and content and paraphrasing.
Reflecting feelings involves
expressing in fresh words the essential feelings, stated or strongly
implied, of a student.
Reflecting content is repeating in fewer and fresher words the
essential ideas of the student.
It is used to clarify ideas that the
student is expressing with difficulty and confusion.
Paraphrasing, which is very similar to reflecting content, is
a method of restating the student's basic message in similar, but
usually in fewer words (Brammer, 1973:90).
Example:
Student:
"I just don't understand.
One minute she
78
tells me to do this and the next minute
to do that."
Instructor:
Example:
"She really confuses you."
"Do you feel this assignment was too difficult?"
If an instructor paraphrases a student statement, then both
this category and category 2 are checked.
5.
Instructor speaks positively of others:
This category
includes instructor statements that he likes someone, that he thinks
someone did
a
good job, that he thinks someone looks good, and that he
thinks someone is smart, kind, funny, etc.
Included only are statements
about people not present in the classroom.
Also statements that merely indicate that the instructor
thinks positively of others are not included.
To be included in this
category, the instructor must actually state his opinion or express
his feelings by using such words as like, good job, smart, considerate,
or pretty.
6.
'
Instructor speaks negatively of others:
This category
includes instructor statements that he dislikes someone, that someone
looks ugly, that someone did a bad job, or that he thinks someone is
dumb, cruel, boring, etc.
Included only are statements about people
-
not present in the classroom.
Also, statements that merely indicate that the instructor
thinks negatively of others are not included.
To be included in this
I
79
category, the instructor must actually state his opinion or express
his feelings by using such words as dislike, bad job, ugly, dumb,
inconsiderate, etc.
7.
Instructor smiling:
This category refers to any instructor
behavior which can be construed as a smile— even a slight smile.
Laughing is considered a smiling behavior.
A smile is considered terminated only when the mouth returns
to a neutral position.
Hence, a continuous smile with varying degrees
of intensity is counted only once.
8.
Student laughter: This category refers to any audible
student behavior which can be construed as laughter and which is in
response to the instructor.
Student laughter is considered terminated when a three-second
period of no laughing follows (see Appendix B, pages 150 and 151,
for a copy of the rating sheet used in the study).
Reliability
Test-retest reliability was established for the Teacher
Behavior and Characteristic Checklist through the following procedure.
First, fifteen students who were volunteers from an Educational
Psychology 208 class were trained by the investigator to count and
I
categorize instructor behaviors according to the Teacher Behavior and
Characteristic Checklist.
, f
This training took place during Winter
Quarter 1975 and included five one-hour sessions and one. one-and-one-
80
half hour session over a period of six weeks.
Both audio and video
tapes were used during the training sessions.
During the sixth ses­
sion, the students viewed a twenty-five minute video tape of an actual
situation and quantified instructor behaviors in accordance with the
checklist.
The video tape depicted the investigator as the instructor
making a presentation and leading a discussion in a regular class
session of a Philosophy of Education five course.
Then ten days after
the first rating, the students once again rated the same tape— eight
students failed to attend the last session.
The reliability with which the students rated the tape was
determined for the nine categories tested (instructor dress and status
categories were not tested) by comparing their ratings on the first
occasion with their ratings on the second occasion.
The results of the
Chi Square statistics used in this analysis appears in Table 3 below.
Table 3
Reliability of Ratings
Rater
X2
Rater
X2
I
8.17
5
.706
2
3.69
6
.796
3
1.45
7
.757
4
1.30
8
.498
81
Validity
The validity established for the Teacher Behavior and Charac­
teristic Checklist is primarily content validity.
Upon completion of
the instrument, three professionals in the field of education examined
the instrument and compared it to the stated goals of the study in
order to determine if the needed data would be adequately collected
and quantified.
Minor revisions were made during the rater training
sessions when there was a need to further describe the behaviors
included in certain categories.
Data Relating to Teacher
Warmth
A modified version of the Scale for Measurement of Counselor
Traits, an instrument developed by Suvak (1966), was administered to
the students included in the study in order to obtain a measure of the
degree to which the students viewed their instructors as warm persons,
i.e., the degree to which they felt liked, respected, and understood by
their instructors.
The instrument consists of twenty-eight five-choice, strongly
agree to strongly disagree scales which require about ten minutes to
complete.
Reliability
The reliability of the Scale for Measurement of Counselor
Traits was determined by both the odd-even and test-retest methods.
82
The odd-even reliability coefficients were computed by obtaining a
sum of first the responses to the odd-numbered items and then a sum of
the even-numbered items.
The author considered the adjusted reliabil­
ity. coefficients which ranged from .790 to .907 to be an adequate
measure of reliability.
Another reliability estimate of the Scale for Measurement of
Counselor Traits was made which involved a test-retest comparison.
Pearson r coefficients derived from this procedure ranged from .139 to
.673, with a mean of .418 for the thirty-two correlations.
All but
the r of .139 reached significance at the .05 level.
Validity
The validity of.the Scale for Measurement of Counselor Traits
was assessed by making a comparison between the Scale for Measurement
of Counselor Traits and the scales developed by .Truax.
Subjects were
asked to evaluate three ten-minute taped counseling interviews on the
Scale for Measurement of Counselor Traits and the Truax scales.
Pearson r correlation coefficients were computed between the Scale for
Measurement of Counselor Traits total content scores and the Truax
scores.
Suvak concluded that these validity coefficients which
ranged from .559 to .843 indicated a high positive relationship
between these two scales.
83
Data Relating to Teacher
Prestige
The Authoritativeness Scale developed by McCroskey (1966:67)
was administered to the students included in the study in order to
obtain a measure of the extent to which they viewed their instructors
with prestige.
Student ratings of their instructors according to this
scale were used for the purpose of determining the extent to which the
students viewed their instructors as prestigious.
■ The instrument consists of twenty-two five choice, strongly
agree to strongly disagree Likert scales and requires approximately
ten minutes to complete.
To obtain estimates of item discrimination, reliability, and
validity, the Authoritativeness scale was used in seven experiments
reported by McCroskey.
Introductions for speakers, which were
developed to represent varying prestige levels, constituted the
independent variable in four of these experiments.
In two experiments,
two versions of a speech advocating federal control of education were
tape-recorded and presented with no information as to the source of
the communication.
documented evidence.
cation.
One form of each speech made extensive use of
The other contained no documentation or qualifi­
In the remaining experiment one group of subjects were
instructed to "identify in your mind the speaker whom you would be
most likely to believe, other things being equal."
Another group was
to imagine the speaker they would be "least likely" to believe.
In
84
each experiment, the subjects completed the Authoritativeness Scale
and also a revised version of the Anderson Authoritativeness Scale.
Reliability
The reliability of the Authoritativeness Scale was determined
by both the split-halves reliability estimate and the Hoyt Internal
Consistency reliability estimate.
six of seven experiments.
These estimates were computed for
The split-halves reliability estimates
ranged from a low of .9,44 to a high of .978.
The Hoyt Internal Con­
sistency Reliability estimate ranged from a low of .946 to a high of
.978.
Validity
The author cited three relevant indications of validity for
the Authoritativeness Scale.
First, the content of the items and the
procedure used in their selection tend to indicate that they are
representative samplings of the universe of items pertaining to the
construct of prestige.
Second, this scale correlates highly with the
Anderson authoritativeness scale making it appear that it measures
primarily the same things as the Anderson scale.
Third, all the
hypotheses in all but one experiment were confirmed by the scores
derived from these two scales.
85
Data Relating to Teacher
Effectiveness
Data relating to teacher effectiveness included student
responses to the Revised Faculty Rating Form, to items 33 and 85 of
the Student Perceptions Scale, and to teacher self-ratings.
The Revised Faculty Rating Form, developed by Miller and
Guinouard (1966:2) at Montana State University, is a thirty-five item
questionnaire which provides for student evaluation of teachers.
The
preliminary form of this instrument consisted of sixty-one statements
which were obtained from three sources:
(I) other rating forms,
(2) suggestions of Montana State University faculty members, and
(3) the experimental literature.
During the final week of Winter Quarter 1966, the preliminary
form of sixty-one items was given to students for rating forty-one
teachers in sixty-three classes.
These teachers, who volunteered to
be rated and whose anonymity was protected, were from the four colleges
of Montana State University.
The level of courses ranged from freshman
through senior and the size of the classes ranged from less than ten
to more than two hundred.
According to the student ratings, this
sample of teachers represent about the 60th percentile in general
teaching ability, and the courses they teach are slightly above average
value.
The following statistical analyses were calculated based on the
ratings of the forty-one teachers:
(I) the men and standard deviation
for each item was figured; (2) correlation among all items was
86
figured; and (3) the inter-item correlations were factor analyzed.
The
special statistical criteria for selecting from the preliminary form
thirty-five items that presently make up the Revised Faculty Rating
Form were: "(I) a standard deviation of 1.0 or higher, (2) a mean
greater than 1.5, (3) inter-item correlations low positive or nega­
tive, (4) factor loadings as one of the top ten items for at least
one of the seven factors, and (5) some items stated negatively as
well as some stated positively (Miller and Guinouard, 1966)."
The ratings of teachers evaluated on the Revised Faculty
Rating Form are compared to the norms prepared from the ratings of the
forty-one teachers who were rated Winter Quarter 1966 and reported in
terms of deciles.
There were not sufficient data to develop norms for
items 29 through.35, which were concerned with student assistants and
.laboratory work.
Since classes that are laboratories are not included
in the present study, items 29 through 35 were not included in the
questionnaire administered to students in this study.
Student responses to items 33 and 85 of the Student Perceptions
Scale were looked at separately and were used as measures of teacher
effectiveness.
helped us:
Item 33 reads, "I believe that the class sessions
I. Not at all; 2. Only slightly; 3. Considerably; 4. A
great deal."
Item 85 reads, "How would you rate the overall value of
this course?
I. poor; 2. fair; 3. good; 4. very good; 5. superior."
87
Teacher Self-Ratings
Also looked at separately and used as a measure of teacher
effectiveness were teacher self-ratings.
Teachers included in the
study responded to the question, "Please rate your effectiveness as a
teacher in the class, __________________ .
I. Very low; 2. Low;
3. Average; 4. High; 5. Very high."
METHOD OF COLLECTING DATA
At the beginning of Spring Quarter 1975, thirty-six students
from three sections of Educational Psychology 208 were trained by the
investigator in the use of the Teacher Behavior and Characteristics
Checklist.
The training consisted of two one-and-a-half hour sessions
on two successive days.
During the second session, the students rated
the same twenty-five minute tape used in the validation of the Teacher
Behavior and Characteristics Checklist in order to assess the effective
ness of the training.
Each student was then randomly assigned an instructor whose
classes they attended three times during the quarter— once toward the
beginning, once toward the middle, and once toward the end— for the
purpose of counting and categorizing teacher behaviors according to the
Teacher Behavior and Characteristics Checklist.
They visited the
teachers' classrooms a fourth time to administer to the students the
Student Perceptions Scale and also to ask the students what grade they
88
expected to receive for the course.
These questionnaires required
approximately thirty minutes for the students to complete.
At the end of the quarter, the investigator contacted each of
the teachers in order to obtain their self-evaluations as to the
effectiveness of their teaching in the courses included in the study.
HYPOTHESES
The purpose of this study was to determine the relationships
among teacher behaviors and characteristics and certain measures of
teacher performance.
In the hypotheses that follow, teacher performance was defined
by five independent measures.
1.
These five measures were:
Teacher warmth as measured by items one through thirty-
three of the Student Perceptions Scale.
2.
Teacher prestige as measured by items thirty-four through
fifty-five of the Student Perceptions Scale.
3.
Teacher effectiveness as measured by items fifty-six
through eighty-three of the Student Perceptions Scale.
4.
Student ratings of the value of the course obtained from
their responses to item
5.
thirty-three of the Student Perceptions Scale.
Student ratings of the value of the course obtained from
their response to item eighty-five
of the Student Perceptions Scale.
The following null hypotheses were developed relating to the
I
89
purpose of this study.
1.
There is no significant difference among the four cate­
gories of grades the students expected to receive for the courses and
the five measures of Teacher Performance obtained from student responses
to the Student Perceptions Scale.
2.
There is no significant difference among the three cate­
gories of teacher self-ratings and the five measures of Teacher
Performance obtained from student responses to the Student Perceptions
Scale.
3.
There is no significant difference among the four cate­
gories of teacher rank as measured by the Teacher Behaviors and Charac­
teristics Checklist
and the five measures of Teacher Performance
obtained from student responses to the Student Perceptions Scale.
4.
There is no significant difference among the two categories
of teacher title as measured by the Teacher Behaviors and Characteris­
tics Checklist
and the five measures of Teacher Performance obtained
from student responses to the Student Perceptions Scale.
5.
There is no significant difference among the three cate­
gories of teacher dress as measured by the Teacher
Behaviors and
Characteristics Checklist and the five measures of Teacher Performance
obtained from student responses to the Student Perceptions Scale.
t
6.
There is no significant relationship among student utter­
ances, teacher positive evaluation, teacher smiling, student laughter.
90.
and teacher acknowledgment of student feelings as measured by the
Teacher Behaviors and Characteristics Checklist and the five measures
of the Teacher Performance obtained from student responses to the
Student Perceptions Scale.
7.
There is no significant difference between the two cate­
gories of teacher negative evaluation as measured by the Teacher
Behaviors and Characteristics Checklist and the five measures of
Teacher Performance obtained from student responses to the Student
Perceptions Scale.
8.
■
There is no significant difference between the two cate­
gories of teacher positive experience as measured by the Teacher
Behaviors and Characteristics Checklist and the five measures of
Teacher Performance obtained from student responses to the Student
Perceptions Scale.
9.
There is no significant relationship among the five
categories of Teacher Performance obtained from student responses to
the Student Performance Scale.
ANALYSIS OF DATA
Due to the nature of the data collected in this study, three
statistical methods of analysis were used, analysis of variance,
the Pearson Correlation, and the Duncan's Test.
91
SUMMARY
This chapter was devoted to a description of the community
and study population, instruments used, hypotheses to be tested,
methodology and types of data collected, and how the data was to be
analyzed.
Data pertaining to the characteristics and behaviors exhibited
by instructors in the classroom, data pertaining to the extent to which
these instructors were perceived by their students as warm and presti­
gious, and data pertaining to their effectiveness as teachers were
gathered on the thirty-three instructors of thirty-three courses
taught during Spring Quarter, 1975, at Montana State University.
Instructor characteristics and behaviors were counted and
categorized in accordance with the Teacher Behavior and Characteristic
Checklist, an instrument developed by the investigator.
It includes
eleven categories which are divided into two classes, Measures of
Teacher Prestige and Measures of Teacher Warmth.
The categories in
Class I are:
status and credits, dress, and the relation of positive
experiences.
Class II is made up of the following categories:
student
utterances, positive evaluation of student utterances, negative evalua­
tion of student utterances, acknowledgment of student feelings and
opinions, teacher speaks positively of others, teacher smiling, and
student laughter.
Chapter 4
ANALYSIS AND RESULTS
The analysis and results of this study are presented in this
chapter under the headings of analysis of data and discussion of
results.
Hypotheses six and nine were tested at the .01 level of
significance using the Pearson correlation.
The other seven hypotheses
were tested at the .05 level of significance using the least-square
means analysis and the Duncan's Test.
For these seven hypotheses, an
analysis of variance was used to compare the five categories of teacher
performance with seven different treatments.
The F values computed
for the five categories of teacher performance were found to be
significant beyond the .05 level for each of these hypotheses.
This
significance could be explained by the difference in the characteris­
tics of the instruments used to obtain the five measures of teacher
performance.
For example, whereas the raw scores obtained as measures
of teacher warmth could range from one to thirty-three, the raw scores
obtained as measures of course value— item 33— could range from only
one to five.
Thus, it was felt that no further statistical analysis of
these results was necessary,
An analysis of variance was also used to compare the treatment
and the interaction among the treatments and teacher performance for
these, seven hypotheses.
Only when the computed F values were
93
significant beyond the .05 level was a Duncan's Test then applied.
Data was collected and processed on a total of thirty-three
instructors representing the five academic colleges on the Montana
State University campus (see Table 2, page 72).
ANALYSIS OF DATA
.Null Hypothesis I
Null hypothesis I states:
There is no significant difference
among the four categories of grades the students expected to receive
for the courses and the five measures of Teacher Performance obtained
from student responses to the Student Perceptions Scale.
In testing the null hypothesis, the grades students expected
to receive for the courses were divided into five categories— A, B, C,
D and F, and 0 for those that did not respond to the question.
Table 3, page 94, presents the least-square means and analysis
of variance for each of the five categories of grades and the five
measures of teacher performance.
Table 4, page 95, presents the results of a Duncan's Test,
which compared the least-square means of Table 3 for the five measures
of teacher performance versus the five categories of expected grades.
Only the findings which were significant beyond the .05 level are
included in the table.
The following is a further description of the results shown y
94
Table 3
Least-Square Means and Analysis of Variance Results in
Comparing Expected Grades and Teacher Performance
Expected Grades
Teacher
Performance
0
67
A
149
B
117
C
380
D & F
149
Total
892
Warmth
78.11
75.58
78.91
84.91
88.88
81.27
Prestige
33.52
29.85
31.35
34.09
35.71
32.90
Effectiveness
72.71
75.01
75.25
76.43
75.47
74.97
Course Value—
Item 23
2.81
3.22
2.93
3.75
.2.47
2.83
Course Value—
Item 85
3.08
3.65
3.37
3.00
2.76
3.17
38.05
41.06
40.23
38.36
37.46
N=
Total
Analysis of Variance
Item
DF
Mean-Squares
F
Instructor
4
448856.00000
4022.584 *
Expected Grade
4
1047.22266
9.386 *
Instructor X
Expected Grade
Remainder
* Significant beyond the
16
574.628906
4435
111.583984
.05 level
5.060 *
95
in Table 4.
Warmth.
Student ratings of instructor warmth were significantly
higher for students expecting to get D ’s and F's than (I) students
expecting to get A*s, (2) students who expected to get B ’s, (3) stu­
dents who expected to get C's, and (4) students who did not respond to
the question.
Table 4
Significant Results of the Duncan's Test for
Expected Grades Versus Teacher Performance
Teacher
Performance
Expected
Grade
Expected
Grade
Warmth
A
when compared to
B, C, D&F, 0
Warmth
B
when compared to
C, D&F
Warmth
C
when compared to
D&F, 0
Warmth
D
when compared to
0
Prestige
A
when compared to
C, D&F, 0
Prestige
B
when compared to
C, D&F, 0
Effectiveness
0
when compared to
A, B, C, D&F
Also, students who expected to get C's rated their instructors
significantly higher on the warmth scale than both students who expected
to get B's and students who expected to get A's; and students who
expected to get B's rated their instructors significantly warmer than
students who expected to get A's.
Students who did not respond to the question rated their instruc­
tors significantly warmer than students who expected to get A's but
96
significantly less warm than students who expected to get B's, C ’s, and
D 1s and F 1s.
Prestige.
Students who expected to get D's and F's rated
their instructors significantly higher on the prestige scale than both
students who expected to get B's and students who expected to get A's.
Students who expected to get C's rated their instructors significantly
higher than both students who expected to get B's and students who
expected to get A's.
Students who did not respond to the question
rated their instructors significantly higher than students who expected
to get A's and students who expected to get B's.
Effectiveness♦
Student ratings of instructor effectiveness
were significantly higher for students who did not respond to the
question than (I) students who expected to get A's, (2) students who
expected to get B's, (3) students who expected to get C's, and
(4) students who expected to get D's and F's.
Course Value— Item 33.
In comparing the least-square means
for course value— itme 33 and the five categories of expected grades,
no difference was found that reached the .05 level of significance.
Course Value— Item 85.
In comparing the least-square means
for course value— item 85 and the five categories of expected grades,
no difference was found that reached the .05 level of significance.
Thus, for the measure of warmth it was found that the higher
the grade that the students expected to receive the lower they tended
97
to rate their instructors; and those students that did not respond
rated their instructors lowest except for those students who expected
to get A's.
The same relationship was found to exist for the measure of
prestige.
Students who expected higher grades rated their instructors
lower; and students who did not respond rated their instructors higher
than students who expected to get A's and B's.
For the measure of teacher effectiveness, no definite pattern
was found between student ratings and the grades they expected to
receive.
Students who did not respond were found to rate their instruc
tors the lowest.
Although none of the results for the two measures of course
value, item 33 and item 85, reached significance at the .05 level, it
was found that, in both cases, the higher the grade that the student
expected to receive the higher he rated the value of the course.
Null Hypothesis 2
Null hypothesis 2 states:
There is no significant difference
between the three categories of teacher self-ratings and the five
measures of Teacher Performance obtained, from student responses to the
Student Perceptions Scale.
In testing the null hypothesisj teachers' self-ratings were
divided into three categories, 3.0, 3.5, and 4.0.
None of the teachers
included in the study gave themselves a rating of I, "very poor," or
98
Table 5
Least-Square Means and Analysis of Variance Results in
Comparing Teacher Self-Ratings and Teacher Performance
Teacher
Performance
___________Teacher Self-Ratings____________
3.0
3.5
4.0
Total
N=
249
74
550
873
Warmth
76.85
87.09
78.53
80.83
Prestige
31.76
34.19
30.96
32.30
Effectiveness
75.04
78.38
74.49
75.97
Course Value—
Item 33
3.04
2.61
2.08
2.88
Course Value—
Item 85
3.43
2.76
3.38
3.12
38.02
41.01
Total
<
38.07
Analysis of Variance
Item
DF
Mean Squares
F
Instructor
4
668516.250000
6036.195 *
Self-rating
2
1474.239746
13.311 *
Instructor X
Self-Rating
8
607.918457
5.489 *
4350
110.751251
Remainder
* Significant beyond the .05 level
99
5, "very good."
Table 5, page 98, presents the least-square means and analysis
of variance for each of the three categories of teacher self-ratings
and the five measures of teacher performance.
Table 6 presents the results of a Duncan’s Test which compared
the least-square means, of Table 5 for the five measures of teacher
performance versus the three categories of teacher self-ratings.
Only
the findings which were significant beyond the .05, level are included
in the table.
Table 6
Significant Results of the Duncan's Test for Teacher .
Self-Ratings Versus Teacher Performance
Teacher
Performance
Teacher
Self-Rating
Teacher
Self-Rating
Warmth
3.5
when compared to
3.0, 4.0
Warmth
4.0
when compared to
3.0
Prestige
3.5
when compared to
3.0, 4.0
Effectiveness
3.5
when compared to
3.0, 4.0
The following is a further description of the results shown
in Table 5.
Warmth.
Teachers who rated themselves at 3.5 received signifi­
cantly higher student ratings on the warmth scale than did teachers who
rated themselves at 4.0 and 3.0.
Teachers who rated themselves at 4.0
100
were rated significantly higher on warmth than teachers who rated
themselves at 3.0.
Prestige.
Teachers who rated themselves at 3.5 received
significantly higher student ratings on the prestige scale than did
teachers who rated themselves at 4.0 and 3.0.
Effectiveness.
Teachers who rated themselves at 3.5 received
significantly higher student ratings on the effectiveness scale than
did teachers who rated themselves at 4.0 and 3.0.
Course Value— Item 33.
In comparing the least-square means
for course value— item 33 and the three categories of teacher selfratings, no difference was found that reached the .05 level of
significance.
Course Value— Item 85.
In comparing the least-square means
for course value— item 85 and the three categories of teacher selfratings, no difference was found that reached the .05 level of
significance.
Thus, it was found that teachers who rated themselves at 3.5
received significantly higher student ratings on the warmth, prestige,
and effectiveness scales than did teachers who rated themselves at 4.0
and 3.0.
Also, teachers who rated themselves at 4.0 received signifi­
cantly higher student ratings on the warmth scale than did teachers who
rated themselves at 3.0.
101
Null Hypothsls 3
Null hypothesis 3 states:
There is no significant difference
among the four categories of teacher rank as measured by the Teacher
Behaviors and Characteristics Checklist and the five measures of
Teacher Performance obtained from student responses to the Student
Perceptions Scale.
In testing the null hypothesis, teacher rank was divided into
four categories— instructor, assistant professor, associate professor,
and professor.
Table 7, page 102, presents the least-square means and analysis
of variance for the four categories of teacher rank and the five
measures of teacher performance.
Table 8, page 103, presents the results of a Duncan's Test,
which compared the least-square means of Table 7 for the four cate­
gories of teacher rank.
Only the findings which were significant
beyond the .05 level are included in the table.
The following is a further description of the results shown
in Table 7.
Warmth.
Student ratings of teacher warmth were significantly
higher for associate professors than both assistant professors and
instructors.
Also, professors received significantly higher student
ratings of warmth than both instructors and assistant professors.
102
Table 7
Least-Square Means and Analysis of Variance Results
in Comparing Teacher Rank and Teacher Performance
Teacher Rank
Teacher
Performance
-N=
Instructor
Assistant
Professor
Associate
Professor
Professor
Total
O
77
272
266
258
873
Warmth
74.48
73.89
82.36
81.52
78.06
Prestige
34.58
29.11
31.44
33.03
32.04
Effectiveness
71.90
71.95
76.77
77.25
74.47
Course Value—
Item 33
3.27
2.98
3.01
2.81
3.02
Course Value—
Item 85
3.58
3.45
3.33
3.16
3.38
37.56
36.28
39.38
39.56
Total
Analysis of Variance
Item
DF
Mean-Squares
F
Instructor
4
887627.250000
8241.445 *
Rank
.3
3121.545410
28.983 *
12
1020.089111
Instructor X
Rank
Remainder
*
'4345
Significant beyond the .05 level
107.7040025
9.471
*
103
Prestige.
Student ratings of teacher prestige were signifi­
cantly higher for professors than both associate professors and
assistant professors.
Associate professors were rated significantly
higher than were assistant professors and instructors received the
highest ratings— significantly higher than those received by assistant
professors, associate professors, and professors.
Table 8
Significant Results of the Duncan's Test for
Teacher Rank Versus Teacher Performance
Teacher
Performance
Teacher
Rank
Teacher
Rank
Warmth
Instructor
when compared to
Associate Professor
Professor
Warmth
Assistant
Professor
when compared to
Associate Professor
Professor
Prestige
Instructor
when compared to
Assistant Professor
Associate Professor
Professor
Prestige
Assistant
Professor
when compared to
Associate Professor
Professor
Prestige
Professor
when compared to
Assistant Professor
Associate Professor
Effectiveness
Associate
Professor
when compared to
Instructor
Assistant Professor
Effectiveness
Professor
when compared to
Instructor
Assistant Professor
Effectiveness.
Student ratings, of teacher effectiveness were
significantly higher for professors than both assistant professors and
104
instructors.
Also, associate professors were rated significantly
higher than both assistant professor and instructors.
Course Value— Item 33.
In comparing the least-square means
for course value— item 33 and the four categories of teacher rank, no
difference was found that reached the .05 level of significance.
Course Value— Item 85.
In comparing the least-square means
for course value:— item 85 and the four categories of teacher rank, no
difference was found that reached the .05 level of significance.
Thus, the findings indicated that, for the measure of teacher
warmth, the higher the rank of the teacher the higher he tended to be
rated by the students.
This relationship was also evident for the
measure of teacher prestige with.one notable exception.
Instructors
received the highest ratings— significantly higher than those received
by assistant professors, associate professors, and professors.
How­
ever, it was also found that professors were rated significantly
higher than both associate professors and assistant professors, and
associate professors were rated significantly higher than assistant
professors.
The findings also indicated that, for the measure of teacher
effectiveness, the higher the rank of the teacher the higher he
tended to be rated by the students.
105
Null Hypothesis 4
Null hypothesis 4 states:
There is no significant difference
among the two categories of teacher title as measured by the Teacher
.
Behaviors and Characteristics Checklist and the five measures of
Teacher Performance obtained from student responses to the Student
Perceptions Scale.
In testing the null hypothesis, teacher title was divided
into two categories, doctor and non-doctor.
Table 9, page 106, presents the least-square means and
analysis of variance for the two categories of teacher title and the
five measures of teacher performance.
Table 10, page 107, presents the results of a Duncan's Test
which compared the least-square means of Table 9 for the two categor­
ies of teacher title.
Only the findings which were significant beyond
the .05 level are included in the table.
The following is a further description of the results shown in
Table 9.
Warmth.
Student ratings of teacher warmth were significantly
higher for teachers with the title of doctor than those with a non­
doctor title.
Prestige.
Student ratings of teacher prestige were signifi­
cantly higher for teachers with the title of doctor than those with
a non-doctor title.
Table 9
Least-Square Means and Analysis of Variance Results
in Comparing Teacher Title and Teacher Performance
Teacher Title
Teacher
Performance
Doctor
Non-Doctor
Total
178
695
873
Warmth .
79.94
78.48
79.21
Prestige
34.21
30.76 .
32.49
Effectiveness
74.55
75.09
74.82
Course Value—
Item 2 3
2.87
2.98
2.93
Course Value—
Item 85
3.12
3.39
3.26
38.94
38.14
N=
Total
Analysis of Variance
Item
DF
Means-Square
F
Instructor
4
782145.000000
6986.801 *
Title
I
453.740967
4.053 *
Instructor X
Title
4
398.960449
3.564 *
4355
111.946030
Remainder
* Significant beyond the .05 level
107
Effectiveness.
In comparing the least-square means for
effectiveness and the two categories of teacher title, no difference
was found that reached the .05 level of significance.
Table 10
Significant Results of the Duncan's Test for
Teacher Title Versus Teacher Performance
Teacher
Performance
Teacher
Title
Warmth
Doctor
when compared to
Non-Doctor
Prestige
Doctor
when compared to
Non-Doctor
Course Value— Item 33.
Teacher
Title
In comparing the least-square means
for course value— item 33 and the two categories of teacher title, no
difference was found that reached the .05 level of significance.
Course Value— Item 85.
In comparing the least-square means
for course value— item 85 and the two categories of teacher title, no
difference was found that reached the .05 level of significance.
Thus, for both the measures of teacher warmth and teacher
prestige, student ratings were significantly higher for teachers with
the title of doctor than those with a non-doctor title.
Null Hypothesis 5
Null hypothesis 5 states:
There is no significant difference
among the three categories of teacher dress.as measured by the Teacher
108
Table 11
Least-Square Means and Analysis of Variance Results
in Comparing Teacher Dress and Teacher Performance
Teacher Dress
Teacher
Performance
I.0-1.9
2.0-2.9
3.0-4.0
Total
249
223
265
747
Warmth
75.78
82.67
77.80
78.74
Prestige
31.01
32.33
29.65
30.10
Effectiveness
72.20
76.79
77.35
75.45
Course Value—
Item 33
2.85
2.91
3.12
2.96
Course Value—
Item 85
3.27
3.23
3.51
3.34
37.02
39.59
38.29
N=
Total
Analysis of Variance
Item
DF
Means-Square
F
Instructor
4
1031926.500000
9533.328 *
Dress
2
1979.422372
18.287 *
Instructor X
Dress
8
872.572266
3720
109.244080
Remainder
* Significant beyond the .05 level
8.061
*
109
Behaviors and Characteristics Checklist and the five measures of
Teacher Performance obtained from student responses to the Student
Perceptions Scale.
In testing the nullvhypothesis, teacher dress was divided into
three categories— 1.0-1.9» 2.0-2.9, and 3.0-4.0.
Table 11, page 108, presents the least-square means and analy­
sis of variance for the three categories of teacher dress and the five
categories of teacher performance.
Table 12 presents the results of a Duncan's Test which compared
the least-square means of Table 11 for the three categories of teacher
dress.
Only the findings which were significant beyond the .05 level
are included in the table.
Table 12
Significant Results of the Duncan's Test for
Teacher Dress Versus Teacher Performance
Teacher
Performance
Teacher
Dress
Teacher
Dress
Warmth
2.0-2.9
when compared to
Warmth
3.0-4.0
when compared to
I.0-1.9
Prestige
2.0-2.9
when compared to
3.0-4.0
. 2.0-2.9
.when compared to
I.0-1.9
Effectiveness
1.0-1.9, 3.0-4.6
The following is a further description of the results shown
in Table 11.
HO
Warmth.
Student ratings of teacher warmth were significantly
higher for teachers in the dress category 2.0-2.9 than teachers .in'
the 1.0-1.9 category and teachers in the.3.0-4.0 category.
Also,
teachers in the 3.0-4.0 category were rated significantly higher than
teachers in the I.0-1.9 category.
Prestige.
For the measure of teacher prestige, teachers in
the 2.0-2.9 category received significantly higher student ratings than
teachers in the 3.0-4.0 category.
Effectiveness.
For the measure of teacher effectiveness,
teachers in the 2.0-2.9 category were rated significantly higher by
the students than teachers in the I.0-1.9 category.
Course Value— Item 33.
In comparing the least-square means
for course value— item 33 and the three categories of teacher dress,
no difference was found that reached the .05 level of significance.
Course Value— Item 85.
In comparing the least-square means
for course value— item 85 and the three categories of teacher dress,
no difference was found that reached the .05 level of significance.
Thus, for the measure of warmth, teachers in the 2.0-2.9
category were rated highest followed by teachers in the 3.0-4.0
category and then by teachers in the I.0-1.9 category.
Also, for the measure of prestige, teachers in the 2.0-2.9
category were rated highest; however, only when compared to teachers
Ill
in the 3.0-4.0 category did the difference reach significance beyond
the .05 level.
For the measure of effectiveness, teachers in the
2.0- 2.9 category were rated significantly higher than teachers in the
1.0- 1.9 category.
Null Hypothesis 6
Null hypothesis 6 states:
There is no significant relation­
ship among student utterances, teacher positive evaluation, teacher
smiling, student laughter, and teacher acknowledgment of student
feelings as measured by the Teacher Behaviors and Characteristics
Checklist and the five measures of Teacher Performance obtained from
student responses to Student Perceptions Scale.
Table 13, page 112, presents the correlation coefficients for
the five categories of teacher behaviors and characteristics versus
the five categories of Teacher Performance.
The null hypothesis that the correlation coefficients were 0
was tested at the .01 level of significance.
Negative relationships, significant at the .01 level, were
found to exist between (I) student ratings of teacher warmth and
student utterances, (2) student ratings of teacher warmth and teacher
positive evaluation, (3) student ratings of teacher warmth and teacher
acknowledgment of,student feelings, (4) student ratings of teacher
warmth and teacher smiling,, (5) student ratings of teacher effectiveness
and student utterances, (6) student ratings of teacher effectiveness
112
and teacher positive evaluation, (7) student ratings of teacher
effectiveness and teacher acknowledgment of feelings, and (8) student
ratings of teacher effectiveness and teacher smiling.
Table 13
Correlation Coefficients for Teacher Behaviors and
Characteristics Versus Teacher Performance
-.22*
-.26*
-.19*
-.15*
<r
O
I
Prestige
-.00
-.06
-.05
Effectiveness
-.26*
-.26*
-.34*
Course Value—
Item 33
CO
O
I
O
00
00
O
Course Value—
Item 85
-.02
-.03
.04
§
Warmth
O
Student
Laughter
i
Teacher
Smiling
I
O
VO
Teacher
Acknowledg­
ment of
Student
Feelings
I
O
VO
Teacher
Positive
Evaluation
I
O
Ui
Student
Utterances
I
O
CO
Teacher
Performance
-.19*
.19*
* Significant at .01 level
A positive relationship, significant at the .01 level, was
found to exist between student ratings of teacher effectiveness and '
student laughter.
Null Hypothesis 7
Null hypothesis 7 states:
There is no significant difference
between the two categories of teacher negative evaluation as measured
113
by the Teacher Behaviors and Characteristics Checklist and the five
measures of Teacher Performance obtained from student responses to the
Student Perceptions Scale.
In testing the null hypothesis, teacher negative evaluation
was divided into two categories— 0 and 1-8.
Table 14, page 114, presents the least-square means and
analysis of variance for the two categories of teacher negative evalua­
tion and the five measures of Teacher Performance.
Since the F scores for the group x categories' totals were
not significant at the .05 level, a Duncan's test was not conducted.
Null Hypothesis 8
Null hypothesis 8 states:
There is no significant difference
between the three categories of teacher positive experiences as
measured by the Teacher Behaviors and Characteristics Checklist and
the five measures of Teacher Performance obtained from student
responses to the Student Perceptions Scale.
In testing the null hypothesis, teacher positive experiences
was divided into three categories— 0, 1-6, and 11+.
Table 15, page 115, presents the least-square means and
analysis of variance for the three categories of teacher positive
experiences and the five measures of teacher performance..
Table 16, page 116,, presents the results of a Duncan's Test,
which compared the least+square means of Table 15 for the three
114
Table 14
Least-Square Means and Analysis, of Variance Results
in Comparing Teacher Negative Evaluation and
Teacher Performance
Performance
H
I
Teacher Negative Evaluation
0
N=
383
490
Warmth
77.63
Prestige
31.43
'■ '
74 .1 1 .
Effectiveness
Course Value—
Item 33
Course Value—
Item 85
Total
'
2'SG'
Total
'
■■■ ,
;
:: V ;.
3.26
37.86
79.67
78.65
31.49
31.45
75.66
. 74.88
.3.05 .
2.95
3.40 ..
3.33
38.66
-
Analysis of Variance
Item
DF
Means-Square
F
Instructor
4
1179252.000000
10520.687 *
Negative Evaluation
I
683.693604
6.100 *
Instructor X Nega­
tive Evaluation
4
185.598022
1.656
435
112.088852
Remainder
* Significant beyond the .05 level
115
Table 15
Least-Square Means and Analysis of Variance Results
in Comparing Teacher Positive Experiences and
Teacher Performance
Teacher Positive Experiences
Teacher
Performance
0
1-6
11+
Total
420
.347
106
873
Warmth
78.70
77.42
83.53
79.88
Prestige
31.93
29.24
36.86
32.68
Effectiveness
73.70
75.80
77.34
75.62
Course Value-Item 33
2.98
3.12
2.44
2.85
Course Value—
Item 85
3.35
3.54
2.66
3.18
38.13
37.83
40.57
N=
Total
‘
Analysis of Variance
Item
' '
DF
Means-Square
Instructor
4
Positive Experiences
2
Instructor X Positive
Experiences
8
794.514648
4350
110.354935
Remainder
* Significant beyond the .05 level
t
863769.500000
1585.481934 .
F
7827.191 *
14.367
*
7.200 *
116
categories of teacher positive experiences.
Only the findings which
were significant beyond the .05 level are included in the table.
Table 16
Significant Results of the Duncan’s Test for Teacher
Positive Experiences Versus Teacher Performance
Teacher Positive
Experiences
Teacher
Perfprmance
Teacher Positive
Experiences
11+
when compared to
0, 1-6
Prestige
0
when compared to
1-6
Prestige
11+
when compared to
Ef fectiveness
1-6
when compared to
Effectiveness
11+
when compared to
Warmth
0, 1-6
0
1-6, 0
The following is a further description of the results shown in
Table 15.
Warmth.
Student ratings of teacher warmth were significantly
higher for teachers in the 11+ category of teacher positive experiences
than teachers in the 1-6 category and teachers in the 0 category.
Prestige.
Student ratings of teacher prestige were signifi­
cantly higher for teachers in the 11+ category of teacher positive
experiences than teachers in the 1-6 category and teachers in the 0
r
category.
Also, teachers in the 0 category were rated significantly
higher on the prestige scale than teachers in the 1-6 category.
117
Effectiveness.
Student ratings of teacher effectiveness were
significantly higher for teachers in the 11+ category of teacher
positive experiences than teachers in the 1-6 category and teachers in
the 0 category.
Also,- teachers in the 1-6 category were rated signifi­
cantly more effective than teachers in the 0 category.
Course Value— Item 33.
In comparing the least-square means
for course value— item 33 and the three categories of teacher positive
experiences, no difference was found that reached the .05 level of
significance.
Course Value— Item 85.
In comparing the least-square means
for course value— item 85 and the three categories of teacher positive
experiences, no difference was found that reached the .05 level of
significance.
Thus, for the measure of teacher warmth teachers who related
the most positive experiences were rated the warmest by the students.
Teachers who related the most positive experiences were also rated the
most prestigious.
Although it was found that teachers who did not
relate any positive experiences were rated more prestigious than
teachers in the 1-6 category.
For the measure of teacher effectiveness, teachers who
related the most positive experiences were again rated the highest by
students.
The findings indicated that the more the teacher spoke of
his positive experiences the higher he was rated on the.effectiveness
118
scale.
Null Hypothesis 9
Null hypothesis 9 states:
There is no significant relationship
among the five categories of Teacher Performance obtained from student
responses to the Student Performance Scale.
In testing the null hypothesis, teacher performance was divided
into five categories— teacher warmth, teacher prestige, teacher effec­
tiveness, teacher course value— item 33, and teacher course value— item
85.
Table 17 presents the correlation coefficients for the five
categories of teacher performance versus the five categories of Teacher
Performance.
Table 17
Correlation Coefficients for Teacher Performance
Versus Teacher Performance
Teacher
Performance
Warmth
Prestige
Effec­
tiveness
Course Value
Item 33
X
.47*
.48*
. -.25*
-.36*
-.38*
— •47*
-.04
-.12
.47*
X
.11
Effectiveness
.48*
.11
X
Course Value—
Item 33
-.25*
Course Value
Item 85
-.36*
-.47*
g
*
Prestige
I.
Warmth
Teacher Performance
* Significant beyond .01 level
-.04
-.12.
X
.53*
Course Value
Item 85
.53*
X
119
The null hypothesis that the correlation coefficients were 0
was tested at the .01 level of significance.
Positive relationships, significant at the .01 level, were
found to exist between (I) student ratings of teacher warmth and
teacher prestige, (2) student ratings of teacher warmth and effective­
ness , and (3) student ratings of course value on item 33 and on item
85.
Negative relationships, significant at the .01 level, were
found to exist between (I) student ratings of teacher warmth and
course value item 33, (2) student ratings of teacher warmth and course
value item 85, (3) student ratings of teacher prestige and course value
item 33, and (4) student ratings of teacher prestige and course value
item 85.
DISCUSSION OF RESULTS
In comparing the grades that the students expected to receive
for their courses with the way in which they rated their teachers on
five different scales of teacher performance, definite relationships
were noted for the measures of teacher warmth and teacher prestige.
The findings indicated that for both measures the higher the grades
that the students expected to receive the lower they tended to rate
their instructors.
The findings related to the measure of teacher
warmth are somewhat unexpected.
The literature indicates that people
120
tend to like those that like them; and it may be theorized that
students who think their teachers are going to give them a good grade
would see their teachers as liking them, while students expecting to
receive low grades might tend to think of their teachers as disliking
them.
Thus, it would follow from this view that students expecting to
receive high grades would tend to rate their teachers higher on the
warmth scale than students expecting low grades.
The results of this
study indicated just the opposite.
It might also be expected that students expecting high grades
would tend to value the course more than students expecting low grades.
Although none of the results for the two measures of course value
reached significance at the .05 level, it was found that, in both
cases, the higher the grade the students expected to receive the
higher they rated the value of the course.
The findings relating teacher self-ratings to the five measures
of teacher performance showed that teachers who saw themselves as being
somewhere between "average" and "above average" in effectiveness were
rated higher than the other teachers for the measures of warmth,
prestige, and effectiveness. .Also, teachers who rated themselves as
"above average" in effectiveness were seen by their students as
being warmer than teachers who saw themselves as "average."
It is
interesting to note that none of the teachers included in the study
rated themselves as either "very good" or "very poor."
121
When teacher rank was compared to the five measures of teacher
performance, a positive relationship was found to exist between teacher
rank and student ratings of teacher warmth.
Also, the higher the
rank of the teacher the more students tended to view him with pres­
tige— with one notable exception.
ratings.
Instructors received the highest
This exception might be explained by the fact that both of
the instructors included in the study were young, and possibly their
appointments as instructors were viewed by students as prestigious
achievements.
The findings that teacher rank and prestige appear to
be positively related was expected and appears to be in agreement with
the results of studies by Haiman (1949), Mausner (1953), Goldberg and
Iverson (1965), and Kelman and Hofland (1953), which indicated that
!
information about a sources' experience, training, and competence as
reported by other people seems relevant to the amount of prestige
attributed to him.
With this in mind, the results showing a positive
relationship between teacher rank and prestige appears to be supportive
of the studies by Sarason and Minard (1963), Krumboltz (1968), Monton
(1965), et. al., which provided evidence that people of high prestige
are more effective than people of low prestige in shaping the opinions
of others.
Although none of the results for the two measures of course
value reached significance at the .05 level, it is interesting to note
•
.•
.
.
that, in both cases, the higher the rank of the teacher the lower
122
student ratings of the value of his course tended to be.
It may be
that teachers younger and of lower academic rank approach the teaching
task with greater vigor and ambition.
Concerning teacher title, teachers having a doctor’s degree
were seen by the students as warmer and more prestigious than those
that did not.
These findings were not unexpected.
The finding of a
positive relationship between teacher title and prestige is consistent
with the findings of Haimon (1949), Bochner and Insko (1966), and
Strong and Nixon (1971b), which suggest that an individual's title is
positively related to the extent to which he is held in esteem.
When
comparisons were made between teacher dress and the five measures of
teacher performance, the results showed that teachers who tended to
dress more in ties and dress pants, sports coats and dress pants, and
suits were seen by students as being more warm than both, teachers who
dressed more casually and teachers who wore only sports coats and
suits.
The significant findings for the measure of prestige indicated
only that teachers who tended to dress more in ties and dress pants,
sports coats and dress pants, and suits received higher student ratings
than teachers who wore only sports coats and suits.
If individuals
who wear only sports coats and suits are considered best dressed, then
the results in regard to the measure of prestige are not supportive
of the findings of Lefkowitz, Blake, and Mowton (1955), Mills and
Aronson (1965), and Sarason and Menard (1963), which suggest that the
123
better an Individual dresses the more he is likely to be seen as
prestigious.
However, these results are consistent with the findings
of these studies if one is considered best-dressed if he wears sports
coats and suits some of the time and dresses more casually some of the
time.
The results also showed that teachers who tended to dress
more in ties and dress pants, sports coats and dress pants, and suits
were seen by students as being more effective than teachers who
dressed more casually.
This result is consistent with the findings
of the studies just cited which also indicate that dress is positively
related to one's ability to produce behavior change and opinion change
in another.
When student utterances, teacher positive evaluation, teacher
smiling, student laughter, and teacher acknowledgment of student
feelings were compared with the five measures of teacher performance,
several interesting results were obtained which appear to be inconsis­
tent with some of what the literature indicates.
Significant negative
relationships were found to exist between teacher warmth and teacher
smiling, and between teacher warmth and teacher positive evaluation.
However, the results of studies by Krumboltz, Varenhorst, and Thoreson
(1967), Strong and Nixon (1971), and Schmidt and Strong (1971) indicate
that smiling conveys warmth.
Also, such verbal responses as ."mmm-hmm,"
"good," "okay," "yeah," etc., which were considered in this study to
/
124
be indicators of teacher positive evaluation, might also be construed
as indicators of warmth (Krasner (1958).
Negative relationships were
also found to exist between teacher smiling and teacher effectiveness
and between teacher positive evaluations and teacher effectiveness.
Yet, evidence from several studies supports the hypothesis that social
reinforcement is more effective when it comes from a warm individual
than when it comes from a cold individual (Ferguson and Buss, 1960;
Sapolsky, 1960; and Krasner, 1958).
The thirty studies reported by
Krasner (1958) also provide evidence that positive evaluations like
"imnm-hmm," "good," "okay," "yeah," etc. are effective reihforcers.
A possible explanation for these apparent inconsistencies is
provided by the hypothesis that a teacher can behave too warmly and
thus lose his effectiveness as a reinforcer.
It appears that behaviors
maintained by social reinforcers are responsive to a condition of
relative satiation for such reinforcers (Gerwitz and Baer, 1958; Simkins,
1961).
Also, the results of several studies suggest that a source
might be more effective in producing behavior change in another by
expressing different orders of warmth and/or anger (Johnson, 1971;
Aronson and Linder, 1965;
Sigall and Aronson, 1967; and Deutsch,
Epstein, Canavan, and Gumpert, 1967).
One could conclude that a
teacher who expresses invariant warmth toward his students and always
evaluates their responses in a positive way does not discern or
discriminate very well, or that it.is "just his style" and therefore
125
it is probably not very meaningful to be liked and/or praised by such
a person.
Although the results of the comparison made between teacher
negative evaluation and the five measures of teacher performance were
not statistically significant, it is noteworthy that the trends
were in a direction that is supportive of the above explanation.
For
all five measures of teacher performance, teachers who at times
evaluated student responses negatively received higher student ratings
than teachers who never negatively evaluated student responses.
When the number of times teachers spoke of their positive ex­
periences to their classes was compared to the five measures of teacher
performance, the significant findings indicated that the more a teacher
related his positive experiences the higher he tended to be rated in
terms of warmth and effectiveness and also, with one exception, in
terms of prestige.
The results for the measure of prestige appears to
be supportive of the findings of Haiman (1949), Mausner (1953),
Goldberg and Iverson (1965), and Kelman and Hovland (1953), which
showed that information about a source's experience, training, and
competence as reported by other people seems relevant to the amount of
prestige that is attributed to him.
Although it was found that
teachers who did not relate any positive experiences were rated more
I. ■
prestigious than teachers in the 1-6 category, a possible explanation
for this finding Is that a teacher who discloses little of himself
126
may be more easily perceived by his students as being different from
them and not having the same human qualities as they.
Such a teacher
might be more easily put on a pedestal and viewed with prestige.
This
thinking might also explain why the results of this study showed a
positive relationship between warmth, and the relating of positive
experiences.
The literature indicates that individuals tend to like
others who appear to be like them and also tend to see these others as
liking them.
Thus, teachers who talk little of their personal experi­
ences and are, therefore, perceived by their students as being
different than they might be viewed as less caring and human than
teachers who disclose more of themselves.
The results for the measure
of teacher effectiveness are also consistent with the literature
(Sarason and Minard, 1963; Krumboltz, 1968; Mouton, 1955; et al), which
indicates that people of high prestige are more successful than people
of low prestige in shaping the opinions of others.
When the interrelationships among the five measures of teacher
performance were analyzed statistically, it is interesting that positive
relationships were found to exist between teacher warmth and teacher
prestige, between teacher warmth and teacher effectiveness, and between
student ratings of course value on item 33 and on item 85.
The findings
that teacher warmth and prestige seem to be positively correlated is
consistent with the results of this study cited earlier, that both
teacher warmth and prestige are negatively correlated with, teacher
127
positive evaluations.
The positive relationship that was found to
exist between teacher warmth and teacher effectiveness is in agreement
with the several studies cited previously in this section that
indicated that social reinforcement is more effective when it comes
from a warm individual than when it comes from a cold individual.
Furthermore, the finding that teacher warmth and effectiveness are
positively related is at least not inconsistent with the findings
cited earlier in this section that a negative relationship seems to
exist between teacher warmth and teacher positive evaluation and
between teacher positive evaluation and teacher effectiveness.
Of course, it was expected that student ratings of course value
on item 33 would be positively related to student ratings of course
value on item 85.
The following conclusions were developed upon an analysis of
this study.
I.
In looking at the grades the students expected to receive
for their courses and the way in which they rated their teachers on
the five different scales of teacher performance of the Student
Perceptions Scale, it was found that the higher the grades the students
expected to receive the less they tended to view their teachers as
being warm and prestigious.
For the measure of teacher effectiveness,
no definite relationship was found between student ratings and the
grades they expected to receive.
For the two measures of course value.
128
the results did not reach significance at the .05 level.
2.
As measured by the Student Perceptions Scale, teachers who
rated themselves somewhere between "average" and "above average" in
effectiveness were seen by their students as being more warm, more
prestigious, and more effective than teachers who did not.
Also,
teachers who rated themselves as "above average" in effectiveness
were seen by their students as being warmer than teachers who saw
themselves as "average."
For the two measures of course value, the
results did not reach significance at the .05 level.
3.
As measured by the Student Perceptions Scale and the
Teacher Behaviors and Characteristics Checklist, the higher the rank
of the teacher the more he tended to be seen by his students as being
warm, prestigious, and effective— with, one notable exception.
Teachers
with the rank of instructor were viewed with the more prestige by
students.
For the two measures of course value, the results did not
reach, significance at the ,05 level.
4.
As measured by the Student Perceptions Scale and the
Teacher Behaviors and Characteristics Checklist, teachers having a
doctor’s degree were seen by the students as being warmer and more
prestigious than those who did not have one.
The findings relating to
teacher effectiveness and the two measures of course value did not
reaqh significance at the .05 level.
5.
As measured by the Student Perceptions Scale and the
129
Teacher Behaviors and Characteristics Checklist, teachers who tended
to dress more In ties and dress pants, sports coats and dress pants,
and suits were seen by students as being more warm than both teachers
who dressed more casually and teachers who wore only sports coats and
suits.
Teachers who tended to dress more in ties and dress pants,
.sports coats and dress pants, and suits were viewed with more prestige
than teachers who wore only sports coats and suits, and were seen as
being more effective than teachers who dressed more casually.
For the
two measures of course value, the results did hot reach significance
at the .05 level.
6.
As measured by the Student Perceptions Scale and the
Teacher Behaviors and Characteristics Checklist, teachers were seen
as being warmer and more effective the less they positively evaluated
student responses, the less they acknowledged student feelings, the
less they smiled, and the less their students spoke in class.
Also,
teachers were seen as being more effective the more.student laughter
tended to occur in their classrooms.
The findings relating to teacher
prestige and the two measures of course value did not reach signifi­
cance at the .05 level.
7.
As measured by the Student Perceptions Scale and the
Teacher Behaviors and Characteristics- Checklist, the number of times
a teacher negatively evaluates his students’ responses had no signifi­
cant effect upon student perceptions of teacher warmth, teacher
130
prestige, teacher effectiveness, or the two measures of course value„
8.
As measured by the Student Perceptions Scale and the
Teacher Behaviors and Characteristics Checklist, the more a teacher
related his positive experiences to his class the more he was seen
by his students as being an effective teacher. Also, teachers who
related the most positive experiences were seen as being the most warm
and the most prestigious.
Although teachers who did not talk about any
positive experiences were seen as being more prestigious than those
who talked about only a few.
For the two measures of course value,
the results did not reach significance at the .05 level.
9.
In looking at, the interrelationships among the five cate­
gories of teacher performance as measured by the Student Perceptions
Scale, it was found that the more a teacher was seen by his students
as being warm the more he was also seen as being prestigious and
effective, and the less his course was valued by his students.
was also found that the more
a
It
teacher was seen as being prestigious
the less his course was valued; and, finally, students tended to
answer the two questions used to measure the extent to which they
valued the courses in a consistent wa.y.
The findings indicated that
no significant relationship exists between teacher prestige and teacher
effectiveness or between teacher effectiveness and course value.
131
SUMMARY
The purpose of this chapter is to summarily present and
analyze the data pertaining to this study.
Each of the nine hypotheses
written for the study were described with regard to the degree of
significance of the relationships found among the five measures of
teacher performance and certain teacher behaviors and characteristics.
A discussion of the significant findings in relation to the investiga­
tor's expectations and to what the literature indicates was presented.
Chapter 5 contains a summary of the study with reference to
conclusions and recommendations.
Chapter 5
SUMMARY, CONCLUSIONS, AND RECOMMENDATIONS
Chapter 5 contains a summary of this study, the conclusions,
and the recommendations for further research.
SUMMARY
Adult and higher education is a-topic of vital interest in our
nation today.
The day is here when man, in a world of increased
technology, increased leisure time, and increased social interaction
is finding that education is indeed a life-long process.
Along with
this increased significance of education is a greater need for more
and better adult educators.
Probably no aspect of education has been discussed with
greater frequency, with as much deep concern, or by more educa­
tors and citizens than has that of teacher effectiveness— how
to define it, how to identify it, how to measure it, how to
evaluate it, and how to detect and remove obstacles to its
achievement . . . .
But findings about the competence of teachers
are inconclusive and piecemeal; and little is presently known
for certain about teacher excellence (Biddle and Ellens, 1964:5).
Studies have been done which go far to answer the question as
to the sort of procedures which have been found successful in estab­
lishing good relationships in a classroom and thus contribute to
teacher effectiveness.
However, they do not help the teacher to answer
other questions which arise on actually,'encountering students.
"Have I the correct sort of personality?" . "Will I be able to
win and hold their, attention?
133
It was the intent of this study to determine and describe some
of the characteristics which constitute the "correct sort of person­
ality" and will enable an instructor to win and hold the students’
attention.
Two characteristics which may play an important role in the
teaching-learning process are personal warmth and social prestige.
It
appears that warmth or positive regard from others is a genuine need
of each individual and thus constitutes a powerful social reinforcer.
A teacher who responds to his students in such a way as to contribute
positively to their feelings of self-worth becomes an important person
to those students and capable of modifying their behavior.
The
effectiveness of a teacher's social reinforcement and also his potency
as a social model are further enhanced if he is held in esteem by his
students.
In other words, not only is it rewarding to an individual
to be around another who expresses positive feelings toward him, it
is even more rewarding if that person is of high prestige.
David G. Ryans (1960) notes that it is of interest to consider
the.kinds of behavior people remember about teachers and to raise the
question of relative importance of such remembered characteristics with
respect to behaviors normally assumed to characterize teaching.
Using
a critical-incidents approach in his teacher characteristics study,
Ryan found that most teaching incidents reported (descriptions of actual
observed behavior believed to have contributed to the judgment of
134
superiority in inferiority of the teachers) involved personal or social
teacher behaviors, even though directions had given the judges com­
plete freedom in naming critical incidents.
He then poses and answers
the following question:
Are personal or social characteristics more important than a
teacher’s scholarliness, the teaching procedures followed, unique
demonstrations, or the content taught? One may well doubt they
are more important, but they may be equally important. We ques­
tion why more people often do not mention incidents involving the
teaching learning process per se . . . all of us seek personal
reinforcement and it is in the area of the personal or social
characteristics of teachers and other persons that we best recall
events (Ryans, 1969:72).
It is the contention of the writer that the more a teacher is
held in esteem by his students and the more he contributes verbally
and nonverbally to their feelings of self-worth the more effective
he will be in influencing their behavior.
A teacher is thus in a
position to reinforce learning in his students in either an appropriate
or inappropriate way, either wittingly or unwittingly.
For him to be
an effective teacher he should be aware of those verbal and nonverbal
behaviors which are indicators of warmth and acceptance to his students,
and also those characteristics and behaviors which might cause him to
be held in esteem by his students.
Asch (1946:258) points out:
We look at a person and immediately a certain impression of
his character forms itself in us. A glance, a few spoken words
are sufficient to tell us a story about a highly complex matter.
We know that such impressions form with remarkable rapidity and
with great ease. Subsequent observation may enrich or upset our
135
first view, but we can no more prevent its rapid growth than we
can avoid perceiving a given visual object or hearing a melody.
We also know that this process, though often imperfect, is also
at times extraordinarily sensitive.
It is for these reasons that the characteristics of teacher
warmth and prestige are important aspects of the teaching-learning
process which need investigation.
This study was an attempt to determine the rate of occurrence
of selected characteristics and behaviors of college instructors at
Montana State University which, based on the writer's educational
experiences as a student and research in the areas of personality
theory, communication theory, counseling, and education, seem to be
important factors in the teaching-learning process.
In addition, the problem was to determine the interrelation­
ships among the rate of occurrence of these characteristics and behav­
iors, student perceptions.of teacher warmth and prestige, and teacher
effectiveness.
Thus, three types of data were produced by this study.
Teacher behaviors and characteristics were counted and categorized in
accordance with the Teacher Behaviors and Characteristic Checklist,
an instrument developed by the investigator.
Student perceptions of
teacher warmth, teacher prestige, and teacher effectiveness were
measured by the Student Perceptions Scale.
Finally, student ratings
of course value, teacher self-ratings, and the grades students expected
to receive for the course were obtained.
136
The focus of the literature review was upon three areas:
(I) warmth and its effect upon learning; (2) behavioral cues of
warmth; and (3) prestige and its affect upon learning.
The summary of the literature relating to warmth and its
affect upon learning indicated that the dimension of warmth in the
behavior of teachers and examiners has a major influence on the
behavior of those with whom they interact.
Thus, warmth, may be an
important factor in determining a teacher’s effectiveness.
Warmth is an important quality in a person which seems to
carry more weight than others in establishing a view of an individual’s
personality.
Warmth from others appears to be a genuine need of each
individual and thus constitutes a powerful social reinforcer.
Many
studies have demonstrated the effectiveness of minimal verbal and non­
verbal cues of a teacher or an examiner in conditioning a subject’s
verbal behavior and in some cases nonverbal behavior.
Some of the
cues which can be construed as indicants of warmth or approval are ver­
bal responses such as "mmm-hmm," "good," "okay," "yeah," etc.
Non­
verbal behaviors shown to be effective reinforcers are headnods
and smiles.
Evidence from several studies supports the hypothesis that
social reinforcement is more effective when it comes from a warm
individual than when it comes from a cold individual.
Furthermore, it
appears that a hostile experimenter retards learning in comparison to
137
a neutral experimenter.
It may be that a teacher or an examiner can behave too warmly
and thus lose his effectiveness as a reinforcer.
It appears that
behaviors maintained by social reinforcers are responsive to a condi­
tion of relative satiation for such reinforcers.
The results of
several studies suggest that a source might be more effective in pro­
ducing behavior change in another by expressing different orders of
warmth, and/or anger.
The expression of initially negative feelings
toward another followed by the expression of increasingly positive
feelings might be more rewarding to that person than the expression
of invariant warmth..
The literature relating to the specific behavioral correlates
of warmth indicated that warmth can be expressed both verbally and
/
nonverbally, and that the verbal part of a spoken message may have
considerably less effect on whether a listener feels liked or disliked
than certain nonverbal cues exhibited by a speaker.
The results of several studies indicated that looking at
someone while they are speaking, smiling, and leaning forward when
seated, portray warmth and attentiveness to that person.
An individ­
ual's tone of voice and facial expression while addressing someone
can indicate liking for that person.
There is also some evidence that
maintaining a moderately relaxed posture, standing closer to your
partner and facing him while speaking or listening to.him, communicate
138
positive feelings.
Also, making positive statements about other
people may contribute to one’s being perceived by others as a warm
person.
The literature relating the prestige of a source to his
ability to modify the behavior and attitudes of others indicated that
the effectiveness of a communication in producing opinion change
depends to a large degree on the extent to which the communicator is
held in esteem by the recipient of the communication.
Similarly, the
effectiveness of social reinforcement depends a great deal upon the
prestige of the dispenser of that reinforcement.
Information about a source's experience, training, and compe­
tence as reported by other people seems relevant to the amount of
prestige that is attributed to him.
Physical appearance and manner
of dress are also important determinants of the extent to which, one
is perceived as prestigious.
There is much evidence that a person qonfponted with an opinion
from one who has prestige for him will have his reaction to it colored
accordingly.
Many studies in the area of social influence have shown
that people of high prestige are more successful than
prestige in shaping the opinions of others.
people1of low
The results of several
other studies indicated that the effectiveness of social approval in
modifying behavior is positively related to the prestige of the person
delivering the approval.
Also, the individual's potency as a social
139
model is influenced by such, characteristics as competence and intellec­
tualness, economic and vocational status.
There is some evidence that
if a source is seen as an expert, the degree of warmth he exhibits has
little effect upon his influence power.
For the purposes of this study, data pertaining to teacher
behaviors and characteristics exhibited in the classroom and data
pertaining to teacher warmth, teacher prestige, and teacher effective­
ness was gathered on thirty-six teachers at Montana State University
during Spring Quarter 1975.
At the beginning of this quarter, thirty-
six students from three sections of Educational Psychology 208 were
trained by the investigator in the use of the Teacher Behaviors and
Characteristics Checklist, an instrument developed by the investigator.
The training consisted of two one-and-a-half hour sessions on two
successive days.
During the second session, the students rated the
1I
same twenty-five minute tape used in the validation of the Teacher
Behaviors and Characteristics Checklist in order to assess the effec­
tiveness of the training.
Each student was then randomly assigned a
teacher whose classes they attended three times during the quarter—
.
once toward the beginning, once toward .the middle, and once toward the
end— for the purpose of counting and categorizing teacher behaviors
according to the Teacher Behaviors and Characteristics Checklist.
This
instrument.includes eleven categories which are divided into two
classes— Measures of Teacher Prestige and Measures of Teacher Warmth.
140
The categories in Class I are:
positive experiences.
status and credits, dress, and teacher
Class II is made up of the following categories:
student utterances, positive evaluation of student utterances, negative
evaluation of student utterances, acknowledgment of student feelings
and opinions, teacher speaks positively of. others, teacher smiling,
and s tudent laughter.
The raters visited the teacher's classrooms a fourth time to
administer to the students the Student Perceptions Scale and also to
ask the students what grade they expected to receive for the course.
The Student Perceptions Scale is comprised of three instruments— the
Authoritativeness Scale, a modified version of the Scale for Measure­
ment of Counselor Traits, and the Revised Faculty Rating Form.
At the end of the quarter, the investigator contacted each of
the. instructors in order to obtain their self-evaluations as to the
effectiveness of their teaching in the courses included in the study.
The hypotheses tested in this study were concerned with the
relationships among teacher behaviors and characteristics and five
independent measures of teacher performance.
These five measures were
teacher warmth, teacher prestige, teacher effectiveness, course value—
item 33, and course value— item 85, all of which were measured by the
Student Perceptions Scale.
All hypotheses were tested at either the
.05 or the .01 level of significance.
141
CONCLUSIONS
The following conclusions were developed upon an analysis of
the data of this study.
1.
In looking at the grades the students expected to receive
for their courses and the way in which they rated their teachers on the
five different scales of teacher performance of the Student Perceptions
Scale, it was found that the higher the grades the students expected to
receive the less they tended to view their teachers as being warm and
prestigious.
For the measure of teacher effectiveness, no definite
relationship was found between student ratings and the grades they
expected to receive.
For the two measures of course value, the results
did not reach significance at the .05 level.
2.
As measured by the Student Perceptions Scale, teachers who
rated themselves somewhere between "average" and "above average" in
effectiveness were seen by their students as being more warm, more
prestigious, and more effective than teachers who did not.
Also,
teachers who rated themselves as "above average" in effectiveness were
seen by their students as being warmer than teachers who saw themselves
as "average."
For the two measures of course value, the results did
not reach significance at the .05 level.
3.
As measured by the Student Perceptions Scale and the
Teacher Behaviors and Characteristics Checklist, the higher the rank
,of the teacher the more he tended to be Seen by his students as being
142
warm, prestigious, and effective— with one notable exception.
Teachers
with the rank of instructor were viewed with more prestige by students.
For the two measures.of course value, the results did not reach sig­
nificance at the .05 level.
4.
As measured by the Student Perceptions Scale and the
Teacher Behaviors and Characteristics Checklist, teachers having a
doctor's degree were seen by the students as being warmer and more
prestigious than those who did not have one.
The findings relating to
teacher effectiveness and the two measures of course value did not
reach significance at the .05 level.
5.
As measured by the Student Perceptions Scale and the
Teacher Behaviors and Characteristics Checklist, teachers who tended
to dress more in ties and dress pants, sports coats and dress pants,
and suits were seen by students as being more warm than both teachers
who dressed more casually and teachers who wore only sports coats and
suits.
Teachers who tended to dress more in ties and dress pants,
sports coats and dress pants, and suits were viewed with more prestige
than teachers who wore only sports coats and suits, and were seen as
being more effective than teachers who dressed more casually.
For
the two measures of course value, the results did not reach significance
at the .05 level.
6.
As measured by the Student Perceptions Scale and the
Teacher Behaviors and Characteristics Checklist^ teachers were seen
143
as being warmer and more effective the less they positively evaluated
student responses, the less they acknowledged student feelings, the
less they smiled, and the less their students spoke in class.
Also,
teachers were seen as being more effective the more student laughter
tended to occur in their classrooms.
The findings relating to teacher
prestige and the two measures of course value did not reach signifi­
cance at the .05 level.
7.
As measured by the Student Perceptions Scale and the
Teacher Behaviors and Characteristics Checklist, the number of times a
teacher negatively evaluates his students' responses has no significant
effect upon student perceptions of teacher warmth, teacher prestige,
teacher effectiveness, or course value.
8.
As measured by the Student Perceptions Scale and the
Teacher Behaviors and Characteristics Checklist, the more a teacher
related his positive experiences to his class the more he was seen by
his students as being an effective teacher.
Also, teachers who related
the most positive experiences were seen as being the most warm and the
most prestigious.
Although teachers who did not talk about any posi­
tive experiences were seen as being more prestigious than those who
talked about only a few.
For the two measures of course value, the
results did not reach significance at the .05 level.
9.
In looking at the interrelationships among the five
categories of teacher performance as measured by the Student Perceptions
144
Scale, It was found that the more a teacher was seen by his students
as being warm the more he was also seen as being prestigious and
effective, and the less his course was valued by his students.
It
was also found that the more a teacher was seen as being prestigious
the less his course was valued; and, finally, students tended to
answer the two questions used to measure the extent to which they
valued the courses in a consistent way.
The findings indicated that
no significant relationship exists between teacher prestige and teacher
effectiveness or between teacher effectiveness and course value.
RECOMMENDATIONS
Based upon the findings and conclusions of this study, several
recommendations emerge as appropriate for further investigation.
1.
This investigation, as it related to teacher performance
and student perceptions of teachers, should be replicated on other
college and university campuses, as well as in elementary, secondary,
and vocational school settings.
It is important to know whether the
findings at Montana State University can be applied to other school
settings.
2.
This investigation should be conducted again and modified
to also include an assessment of student perceptions of teacher
performance sometime after the course has ended.
145
3.
It is recommended that more efforts be made to identify
what a teacher does in a classroom in behavioral terms and investigate
the effect of what he does upon the teaching-learning process.
The
results from such, studies would be helpful in the preparation and
evaluation of teachers.
4.
There should be continued research regarding student
perceptions and student learning.
This, too, would provide information
that could aid in the development of instruments for assessing teacher
effectiveness.
5.
It is recommended that teachers be aware of the patterns
of interpersonal communication and the personal characteristics that
they bring to the classroom and their possible effect upon student
perceptions and the teaching-learning process.
APPENDICES
147
APPENDIX A
STUDENT PERCEPTIONS SCALE
Instructions: Please indicate your response to the following items on
the IBM answer sheet provided. Interpret the possible responses as
follows: A - Strongly Agree, B - Agreed., C - Undecided, D - Disagree,
E - Strongly Disagree.
Your responses will be automatically punched.into IBM cards. This
information will be used for research purposes only, and your individ­
ual responses will be kept entirely confidential.
I believe that the instructor:
1.
2.
3.
4.
5.
6.
7.
8.
9.
10.
11.
12.
13.
14.
15.
16.
17.
18.
19.
20.
21.
22.
23.
Exhibited a warm positive feeling toward the students.
Accepted us without establishing any conditions.
Cared for us as individuals.
Was willing to share equally our joys and aspirations or our
depressions and failures.
Valued us as persons without judging us by an evaluation of our
behavior.
Held us in high esteem for ourselves regardless of our behavior.
Actively offered advice.
Attempted to determine the meaning and value of our thoughts and
behavior as we saw them.
Indicated that what we do or do not do is important to him.
Saw himself as responsible to us.
Expressed what he really felt and meant.
Remained aloof.
Indicated that there was a considerable discrepancy between what
he may say and what he believes.
Was himself rather than presenting a professional front.
Gave responses that were sincere.
Indicated an expression of his real feelings rather than being
defensive.
Was himself in all of his responses whether these responses were
personally meaningful or trite.
Attempted to feel the same emotions that we felt.
Made remarks that fit in just right with our mood.
Indicated a sensitive understanding of our most obvious feelings.
Was continuously aware of any emotional shift we made.
Listened carefully and intently to what we said.
Responded to our full range of feelings and communicated an
understanding of our very deepest feelings.
I
148
24.
25.
26.
27.
28.
29.
30.
31.
32.
33.
34.
35.
36.
37.
38.
39.
40.
41.
42.
43.
44.
45.
46.
47.
48.
49.
50.
51.
52.
53.
54.
55.
56.
57.
58.
Seemed to be reflecting his own feelings and experiences rather
than those of students.
Displayed a concern for our deeper, more hidden feelings.
Recognized our present feelings.
Readily recognized mistakes he may have made in understanding our
feelings and was willing to change his perception of us.
Communicated back to us an understanding of our every deepest
feeling.
Seemed to ignore our true feelings.
Appeared to understand exactly what we meant.
Told us what he should do
Was capable of diagnosing our problem.
I believe that the class sessions helped us:
1. Not at all
2. Only slightly
3. Considerably
4. A great deal
I respect this teacher's opinion on the topic.
This teacher is not of very high intelligence.
This teacher is a reliable source of information on the topic.
I have confidence in this teacher.
This teacher lacks information on the subject.
This teacher has high status in our society.
I would consider this teacher to be an expert on the topic.
This teacher's opinion on the topic is of little value.
I believe that this teacher is quite intelligent.
The teacher is an unreliable source of information on the topic.
I have little confidence in this teacher.
The teacher is well-informed on this subject.
The teacher has low status in our society.
I would not consider this teacher to be an expert on this topic.
This teacher is an authority on the topic.
This teacher has had very little experience with this subject.
This teacher has considerable knowledge of the factors involved
with this subject.
Few people are as qualified to speak on this topic as this teacher
This teacher is not an authority on the topic.
This teacher has very little knowledge of the factors involved
with the subject. .
This teacher has had substantial experience with this subject.
Many people are much more qualified to speak on this topic than
this teacher.
The students felt free to express their opinions in class.
The instructor told students when they had done a particularly
good job.
The instructor Assigned very difficult reading.
149
59.
60.
62.
63.
64.
65.
66.
67.
68.
69.
70.
71.
72.
73.
74.
75.
76.
77.
78.
79.
80.
81.
82.
85.
The instructor had everything going according to schedule.
The instructor planned the activities of each class period in
detail.
Students argued with one another or with the instructor, not
necessarily with hostility.
The instructor urged students to greater effort.
Students were given a course outline or syllabus to help organize
their learning.
Assigned material was covered in lectures.
The instructor gave constructive criticism of poor work.
Hand-outs, bibliographies, etc. were used.
The instructor sensed.when students needed help.
Material was presented in such.a manner as to dull student
thinking.
Course objectives were vague or uncertain.
Original thinking was demanded of students.
The grading system was clearly explained.
A large amount of preparation was required outside of class.
The instructor helped students more than most teachers do.
It was hard to get credit when credit was due.
Adequate illustrations or examples were used to clarify important
points.
Students were not told what would be covered on examinations.
The amount of work required was appropriate for the credit
received.
The instructor was impatient with the slower students.
The influence of the instructor on the amount students learned
was insignificant.
Students understood the subject at the end of the course.
Students had difficulty arranging conferences with the instructor.
How would you rate the overall value of this course:
1.
2.
3.
4.
5.
Poor
Fair
Good
Very good
Superior
150
APPENDIX B
TEACHER BEHAVIORS AND CHARACTERISTICS CHECKLIST
Instructor's Name __________________ Course _________ _________________
Dress:
Suit ______
Sports coat with dress pants ______
and dress pants ______
tie ______
Shirt and tie
Dress pants with shirt or sweater; no
Other ______
1.
Student Utterances
Student makes a statement
Student replies to teacher's question
A verbal exchange including two or more student statements is
counted as one utterance
2.
Positive Evaluation
Teacher nods during or after student utterance
Teacher smiles during or after student utterance
Teacher says "good," "yeah," "mmm-hmm,," "okay," etc.
Teacher says "That's a good point" or "I agree"
Teacher refers to student by name
Teacher exhibits a "positive evaluative" behavior at least once
during or after a verbal exchange
Teacher paraphrases student.statement or reflects student
feelings (category #5 behaviors)
3.
Negative Evaluation
Teacher shakes head during or after student ^utterance
Teacher says "no," "wrong," "uh-uh," or "That's wrong"
Teacher sighs or takes a deep breath and doesn't smile
Teacher closes eyes or purses lips; doesn't smile
(If any of these behaviors are followed by "Positive Evaluation"
behaviors, i.e. , smiling,. mark category #2 and not this
category.)
4.
Student Feelings and Opinions
Teacher says "You feel . . . ."
Teacher says "What are your reactions?"
Teacher says "Do you have any comments?"
151
5.
Speaks Positively of Others
Teacher says he likes someone
Teacher says someone did a good job
Teacher says someone looked good
Teacher says someone is smart, kind, funny, etc
6.
Speaks Negatively of Others
Teacher says he dislikes someone
Teacher says someone did a bad job
Teacher says someone looked bad
Teacher says someone is dumb, unkind, etc.
7.
Relates Positive
Teacher talks
Teacher talks
Teacher, talks
Teacher talks
8.
Instructor Smiles
9.
Student Laughter
Experiences
about jobs or positions he’s held
about places he's visited
about personal achievements
about an honor he's received
152
APPENDIX C
November 3, 1975
Thank you for participating in my dissertation study Spring
Quarter, .1975. If you recall, observers sat in on three of your class
sessions, and your students completed a teacher evaluation question­
naire.
The last piece of information I need to complete the study is
your answer to the following question. Your response will be held in
confidence. A self-addressed envelope is enclosed for your convenience.
Please rate your effectiveness as a teacher in. the class, __________ ."
I.
Very low
2.
Low
3.
Average
4.
High
5.
Very high
Please check here if you are interested in receiving the results df
this studyj~~ | .
Sincerely,
Doug Smith
LITERATURE CITED.
LITERATURE CITED
Argyle, M. and A. Kendon. 1967. The experimental analysis of social
performance. Advances in Experimental Social Psychology, 3:55-98.
Aronson, E. and D. Linder. 1965. Gain and loss of esteem as determin­
ants of interpersonal attractiveness. Journkl of Experimental
Social Psychology, 1:156-171.
Aronson, E., J. Turner, and J. M. Carlsmith. 1963. Communicator
credibility and communication discrepancy as determinants of
opinion change. The Journal of Abnormal and Social Psychology,
67:31-36.
Asch, S. E. 1946. Forming impressions of personality.
of Abnormal and Social Psychology, 41:258-290.
The Journal
Bandura, A. and R. H. Walters. 1963. Principles of Behavior Modifi­
cation. New York: Holt, Rinehardt, and Winston.
Barron's profiles of American colleges: An in-depth study of Montana
State University. 1974. Barron's Educational Series, Inc..
Woodbury, N.Y., 5.
Bayes, M. A. 1972. Behavioral cues of interpersonal warmth.
of Consulting and Clinical Psychology, 9:33-39.
Journal
Bergin, A. E. 1962. The effect of dissonant persuasive communications
upon changes in a self-referring attitude. Journal of Personality,
30:423-438.
Biddle, B. and J. Ellens. 1964. Additions to a bibliography on the
present status of role theory (No. 2). Journal of Social
Psychology, 31:39-60.
Bochner, S. and C. A. Insko. 1966. Communicator discrepancy source
credibility, and opinion change. Journal of Personality and
Social Psychology, 4:614-621.
Brammer, L. M. 1973. The Helping Relationship Process and Skills.
Englewood Cliffs, N.J.: Prentice-Hall, Inc.
Browning, G. J. 1965. An analysis of the-effects of therapist pres­
tige and levels of therapist prestige|and levels of interpretation
on client response in the initial phase of psychotherapy. Disser­
tation Abstracts, 4803-4804.
155
Byrne, D. 1963. Assessing personality variables and their alteration.
In P. Worchal and D. Byrne (Eds.) Personality Change. New York:
John Wiley & Sons, Inc.
Cientat, V. J. 1959. Surreptitious modification of verbal behavior
during class discussion. Psychological Reports, 5:648.
Das, J. P. 1960. Prestige effects in body-sway suggestibility.
Journal of Abnormal and Social Psychology, 61:487-488.
The
Dawson, J. E., L. A. Messe, and J. L. Phillips. 1972. Effect of
instructor-leader behavior on student performance. Journal of
Applied Psychology, 56(5):369-376.
Deutsch, M., Y. Epstein,D. Canavan, and P . Crumpert. 1967. Strategies
of inducing cooperation: An experimental study. Journal of
Conflict Resolution, 11:345-361.
Ekman, P., L. Krasner, and L. Ullmann. 1963. Interaction of set and
awareness as determinants of response to verbal conditioning.
Journal of Abnormal and Social Psychology, 66(4):387-389.
Ferguson, D. C. and A; H. Buss. 1960. Operant conditioning of hostile
verbs in relation to experimenter and subject characteristics.
Journal of Consulting Psychology, 24:324-327.
Ford, D. H. and H. B. Urban. 1963. Systems of Psychotherapy:
Comparative Study. New York: John Wiley arid Sons, Inc.
Gage, N. L. 1972. Teacher Effectiveness and Teacher Education.
Alto, California: Pacific Books.
A
Palo.
Gerwitz,. J. L. and D. M. Baer. 1958. The effect of brief social
deprivation on behaviors for a social reinforcer. The Journal
of Abnormal and Social Psychology, 56:49-56.
Gerwitz, J. L. and D . M. Baer. 1958. Deprivation and satiation of
social reinforcers as drive conditions.. Journal of Abnormal and
Social Psychology, 57:165-172.
Getzels, J. W. and P. W. Jackson. 1963, The teacher's personality and
characteristic. In N. L. Gage (Ed.) Haridbbok of Research on
Teaching. Chicago: Rand McNalley.
156
Guilford, J. P. and B. Fruchter. 1973. Fundamental Statistics in
Psychology and Education. New York: McGraw-Hill.
Goldberg, H. and M. Iverson. 1965. Inconsistency in attitude of high
status persons and loss of influence: An experimental study.
Psychological Reports, 16:673-683.
Haase, R. F. and D. T. Tapper, Jr. 1972. Nonverbal components of
empathic communication. Journal of Counseling Psychology, 19(5):
417-419.
Haiman, F. S. 1949. An experimental study of the effects of ethos and
public speaking. Speech Monographs, 16:192-202.
Hendrix, W. L.
1960.
Harris, T. A. '1973.
Principles of Teaching.
I'm OK— You're OK.
Chicago: Rand McNally.
New York: Harper and Rowe.
Hovland, C. J., N. C. Janis, and H. G. Kelly. 1953. Communication an
Persuasion: Psychological Studies of Opinion Change. Yale Uni-r
versify Press.
James, W. T. A. 1932. A study of the expression of bodily posture.
Journal of General Psychology, 7:405-437.
Johnson, D. W. 1971. Effects of warmth of interaction, accuracy of
understanding, and the proposal of compromises on listener's
behavior. Journal of Counseling Psychology, 18(3):297-216.
Johnson, D. W. 1971. Effects of the order of expressing warmth and
anger on the. actor and the listener. Journal of Counseling
Psychology, 18(6):571-578.
Kelly, H. H. 1950. The warm-cold variable in first impressions of
persons'. Journal of Personality, 18:431-439.
KeIman, H. C. and C. I. Hovland. 1953. Reinstatement of the communi­
cator in delayed measurement of opinion change. The Journal of
Abnormal and Social Psychology, 48(3):327-335.
Knowles, M. 1973. The Adult Learner: A ^eglectqd Species.
Gulf Publishing Company.
Houston:
Krasner, L. 1958. Studies of the conditioning of verbal behavior.
Psychological Bulletin, 55:148-170.
157
Krasner, L., L. P. Ullmann, R. L. Weiss, and B. F. Collins. 1961.
Responsivity to verbal conditioning as a function of three
different examiners. Journal of Clinical Psychology, 17:411-415.
Krasner, L. 1962. The therapist as a social reinforcement machine.
In H. H. Strupp and L. Luborsky (Eds.) Research in Psychotherapy.
Washington, D.C.: American Psychology Association.
Krasner, L. and L. P. Ullmann. Research in Behavior Modification.
York: Holt, Rinehardt, and Winston, Inc., 1965.
New
Krumboltz, J. D., B. B. Varenhorst, and C. E. Thoreson. 1967. Non­
verbal factors in the effectiveness of models in counseling.
Journal of Counseling Psychology, 14:412-418.
Kulp, D. H. 1934. Prestige as measured by single-experience changes
and their permanency. Journal of Educational Research, 27:663672. ■
Lehat-Mandelbaum, B. and D. Kipins. 1973. Leader behavior dimensions
related to students’ evaluation of teacher effectiveness. The
Journal of Applied Psychology, 58(2):250-253.
Lancelot, A., E. Barr, J. F. Torgerson, M. S..Johnson, P. Lynn, J. S.
Walaoord, and D. W. Detts. 1935. Traits and characteristics of
effective teachers. Journal of Educational Research, 18:47-79.
I
Lefkowitz, M., R. R. Blake, and J. S. Mouton. 1955. Status factors
in pedestrian violation of traffic signals. The Journal of
Abnormal and Social Psychology, 51:704-706.
Lorge, I. 1936. Prestige, suggestion, and attitudes.
Social Psychology, 7:402-408.
Machotka, P. 1965. Body movement as communication.
Science Research, 2:33-66.
Journal of
Behavioral
Maslow, A. H. and W. Zimmerman. 1956. College teaching ability,
scholarly activity and personality. Journal of Educational
Psychology, 47:185-189.
Matarazzo, J. D., G. Saslow, A. N. Wiens, M. Weitman, and B. V. Allen.
1964. Interviewer head-nodding and interviewee speech duration.
Psychotherapy, 1:54-63.
158
Matarazzo, J. D., A. N. Wiens, G. Saslow, B. V. Allen, and M. Weitman.
1964. Interviewer "m-hmm" and interviewee speech durations.
Psychotherapy, 1:109-114.
Mausner, B. 1953. Studies in social interaction: III effect of
variation in one partner's prestige on the interaction of
observer pairs. The Journal of Applied Psychology, 37(5):391-393.
McCroskey, J. C. 1966. Scale for measurement of ethos.
Monographs, 33:65-73.
Mehrabian, A. 1968.
2(4):52-54.
Communication without words.
Speech
Psychology Today,
Menard, T. L. 1973. Analysis of the relationship between teacher
effectiveness and teacher appearance. Dissertation Abstracts
International, 33(7A):3394-A.
Mills, J. and E. Aronson. 1965. Opinion change as a function of the
communicator's attractiveness and desire to influence. Journal
of Personality and Social Psychology, 1:173-177.
Miller, A. W. and D. E. Guinouard. 1966. Report on the Revised
.Faculty Rating Form. Montana State University, unpublished.
Oakes, W. F. 1962. Effectiveness of signal light reinforcers given
various meanings on participation in group discussion. Psychologi­
cal Reports, 11:469-470..
Riley, A. M., G. Ryan, and C. A. Lifschitz. 1950. Correlates of
studepts' perceptions of ideal factors in teaching. Journal of
Consulting Psychology, 17:193-201.
Rogers, C. R.
Mifflxn.
1951.
Client-Centered Therapy.
Boston: Houghton-
Rosenbaum, M.. E. and I. F. Tucker. 1962. The competence of the model
and. the learning of imitation and nonimitation. Journal of
Experimental Psychology, 63(2):183-190.
Ryans, D. G. 1960. Characteristics of Teachers: Their Description,
Comparison, and Appraisal. Washington, D.C.: American Council
on Education.
.
159
Ryans, D. G. 1969. Measurement and prediction of teacher effective­
ness. In A. Anastasi (Ed.) Testing Problems in Perspective.
Twenty-fifth anniversary volume of topical readings from the
Invitational Conference on Testing Problems. Washington, D.C.:
American Council on Education, 222-237.
Salzinger, K. 1959. Experimental manipulation of verbal behavior: A
review. Journal of General Psychology, 61:65-94.
Sapolsky,
1960. Effect of interpersonal relationships upon verbal
conditioning. The Journal of Abnormal and Social Psychology, 60:
241-246.
Sarason, I. G. 1958. Interrelationships among individual difference
variables, behavior in psychotherapy, and verbal conditioning.
The Journal of Abnormal and Social Psychology, 56:339-344.
Sarason, I. G. 1962. Individual differences, situational variables,,
and personality research. Journal of Abnormal and Social
Psychology, 65(6):376-380.
Sarason, I. G. and J. Minard. 1963. Interrelationships among subject,
experimenter, and situational variables. The Journal of Abnormal
and Social Psychology, 67(1):87-91.
Schmidt, L. D. and S. R. Strong. 1971. Attractiveness and influence
in counseling. Journal of Counseling Psychology, 18:348-351.
Sigall, H. and E. Aronson. 1967. Opinion change and. the gain-loss
model of interpersonal attraction. Journal of Experimental
Social Psychology, 3:178-188.
Simkins, L. 1961. Effects of examiner attitudes and types of rein­
forcement on the conditioning of hostile verbs. Journal of
Personality, 29:380-395.
Skinner, B. F.
Crogts.
1957.
Verbal Behavior.
New York: Appleton-Century-
Strong, S. R. and D. N. Nixon. 1971. ExpertnesS, attractiveness, and
influence in counseling. Journal of Counseling Psychology, 18(6):
562-570.
Suvak, A. 1966.
University.
Scale for Measuring Counselor Traits.
Unpublished.
Montana State
160
Tedeschi, J. T., T. V. Bonoma, and B. Schlenker. 1972. Influence,
decision, and compliance. In J. T. Tedeschi (Ed.), The Social
Influence Processes,
MONTANA STATE UNIVERSITY LIBRARIES
3 1762 10015624 7
Smith, Douglas N
Relationships among
selected teegiter behav­
iors and characteristics
and studeniperceptions
of teacher warmth,
prestige ...